Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

🚨 Same sales data. Same prompt. ~32% estimated lower cost. I put the exact same analysis through: Claude Opus 4.8 → $0.12 vs. GMI Router → Cost Mode → Gemini 3.7 Flash → ~$0.081 The task wasn't simple: 6 months of sales data across 4 products → trends, product...

altiamkabir's profile picture

Altiam Kabir

66,730 subscribers

32,438 Aufrufe • vor 2 Tagen •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

0 Kommentare

Keine Kommentare verfügbar

Kommentare vom Original-Post werden hier angezeigt

Ähnliche Videos

🚨 GMI estimated ~$0.11 in savings on one 3-step Python coding workflow. I gave GMI Router the same e-commerce codebase and split the work into 3 requests: 👉 Code analysis → Cost Mode → DeepSeek-V4-Flash → ~$0.046 saved 👉 Debugging → Balanced Mode → DeepSeek-V4-Flash → ~$0.042 saved 👉 Implementation + testing → Quality Mode → DeepSeek-V4-Flash → ~$0.019 saved Total: ~$0.11 estimated savings vs Claude Opus 4.8, based on GMI's estimated token usage. But here's what caught my attention: Router selected the same model for all 3 tasks — even though I changed the mode each time. And that's actually the point. GMI Cloud 's Router is cache-aware. When the same model can handle the next request, sticking with it can help preserve cache reuse and reduce cost. There's another reason: These tasks weren't especially difficult. So even in Quality Mode, Router didn't need to move to a heavier model. The result: 1 codebase. 3 tasks. 3 routing modes. 1 lightweight model. ~$0.11 estimated savings. This is why request-level routing is more interesting than simply assigning a different model to every task.
1:11

🚨 GMI estimated ~$0.11 in savings on one 3-step Python coding workflow. I gave GMI Router the same e-commerce codebase and split the work into 3 requests: 👉 Code analysis → Cost Mode → DeepSeek-V4-Flash → ~$0.046 saved 👉 Debugging → Balanced Mode → DeepSeek-V4-Flash → ~$0.042 saved 👉 Implementation + testing → Quality Mode → DeepSeek-V4-Flash → ~$0.019 saved Total: ~$0.11 estimated savings vs Claude Opus 4.8, based on GMI's estimated token usage. But here's what caught my attention: Router selected the same model for all 3 tasks — even though I changed the mode each time. And that's actually the point. GMI Cloud 's Router is cache-aware. When the same model can handle the next request, sticking with it can help preserve cache reuse and reduce cost. There's another reason: These tasks weren't especially difficult. So even in Quality Mode, Router didn't need to move to a heavier model. The result: 1 codebase. 3 tasks. 3 routing modes. 1 lightweight model. ~$0.11 estimated savings. This is why request-level routing is more interesting than simply assigning a different model to every task.

Anthara Fairooz

28,246 Aufrufe • vor 2 Tagen

We’re launching Optima. Now anyone can create a custom benchmark for their use case, leveraging Artificial Analysis’ leading research and platform Building and running benchmarks is difficult. We have distilled Artificial Analysis’ research and experience developing benchmarks into Optima, a new platform for benchmarking models on your own workloads and comparing performance, speed and cost efficiency. Optima allows you to find the best model for your task, or an equally performant alternative to your current setup at 10x lower cost or time per task. We’ve integrated Artificial Analysis' research and experience in benchmarks across the Optima workflow: ➤ Build benchmarks based on your own data and use cases: There are three ways to build a benchmark with Optima. Upload an existing evaluation dataset from your own files or Hugging Face, or import agent traces from platforms including Arize AI, Braintrust and langfuse.com. Install the Optima skill to build a benchmark using context from your coding environment and previous sessions. Or simply describe your use case and provide example inputs and outputs, and Optima will build the benchmark for you ➤ Run across the latest models: Run the same benchmark across leading models in a single click, and keep your leaderboard up to date as soon as new models are released ➤ Bring Artificial Analysis grading to your own benchmark: Evaluate responses against objective rubric criteria or using the same pairwise judging approach used for Artificial Analysis benchmarks including GDPval-AA and AA-Briefcase. For pairwise judging, select your preferred responses from a sample and Optima uses those preferences to rank models across your test set ➤ Compare performance, cost and time efficiency: Optima measures more than model performance. Cost per Task and Time per Task are tracked alongside benchmark scores, with category-level results and support for custom metrics, allowing you to compare the tradeoffs between models for your specific use case Ahead of launch, here are examples questions our beta testers answered with Optima: ➤ Which model can save me 10x the cost without a meaningful decrease in quality for my finance & accounting agent? ➤ Which model best matches the writing style of lawyers for my legal agent? ➤ Which model can best identify different elements in my custom image dataset? Optima is available today. Build your own benchmark at
2:33

We’re launching Optima. Now anyone can create a custom benchmark for their use case, leveraging Artificial Analysis’ leading research and platform Building and running benchmarks is difficult. We have distilled Artificial Analysis’ research and experience developing benchmarks into Optima, a new platform for benchmarking models on your own workloads and comparing performance, speed and cost efficiency. Optima allows you to find the best model for your task, or an equally performant alternative to your current setup at 10x lower cost or time per task. We’ve integrated Artificial Analysis' research and experience in benchmarks across the Optima workflow: ➤ Build benchmarks based on your own data and use cases: There are three ways to build a benchmark with Optima. Upload an existing evaluation dataset from your own files or Hugging Face, or import agent traces from platforms including Arize AI, Braintrust and langfuse.com. Install the Optima skill to build a benchmark using context from your coding environment and previous sessions. Or simply describe your use case and provide example inputs and outputs, and Optima will build the benchmark for you ➤ Run across the latest models: Run the same benchmark across leading models in a single click, and keep your leaderboard up to date as soon as new models are released ➤ Bring Artificial Analysis grading to your own benchmark: Evaluate responses against objective rubric criteria or using the same pairwise judging approach used for Artificial Analysis benchmarks including GDPval-AA and AA-Briefcase. For pairwise judging, select your preferred responses from a sample and Optima uses those preferences to rank models across your test set ➤ Compare performance, cost and time efficiency: Optima measures more than model performance. Cost per Task and Time per Task are tracked alongside benchmark scores, with category-level results and support for custom metrics, allowing you to compare the tradeoffs between models for your specific use case Ahead of launch, here are examples questions our beta testers answered with Optima: ➤ Which model can save me 10x the cost without a meaningful decrease in quality for my finance & accounting agent? ➤ Which model best matches the writing style of lawyers for my legal agent? ➤ Which model can best identify different elements in my custom image dataset? Optima is available today. Build your own benchmark at

Artificial Analysis

130,269 Aufrufe • vor 16 Tagen

the tiebreaker is done. qwen 3.5 27B dense. single RTX 3090. one prompt. zero steering. zero human edits. 1,827 lines across 10 files. 13 minutes. full thinking mode. runs on first load. hermes 4.3 got the same prompt with 2x 3090s and 5x the context it needed. wrote 1,249 lines, left empty files, needed 3 interventions, game was broken on load. same architecture class. same quant. hermes got double the hardware. completely different result. dense wasn't the problem. hermes was. but here's what got me. this model thinks at 27 tok/s. every single token carries 27 billion parameters of reasoning. MoE hit 112 tok/s but only 3B active per token. the dense model is slower and it doesn't matter. watch 13 minutes of autonomous coding on a consumer GPU with zero intervention and tell me speed is what matters. a year ago this wasn't possible. now it runs on hardware you can buy used for $900. no API. no subscription. no cloud. just a 3090 doing what data centers did 18 months ago. full unedited session in the video. every token, every file, every thinking chain. 16 minutes. hit play.
16:04

the tiebreaker is done. qwen 3.5 27B dense. single RTX 3090. one prompt. zero steering. zero human edits. 1,827 lines across 10 files. 13 minutes. full thinking mode. runs on first load. hermes 4.3 got the same prompt with 2x 3090s and 5x the context it needed. wrote 1,249 lines, left empty files, needed 3 interventions, game was broken on load. same architecture class. same quant. hermes got double the hardware. completely different result. dense wasn't the problem. hermes was. but here's what got me. this model thinks at 27 tok/s. every single token carries 27 billion parameters of reasoning. MoE hit 112 tok/s but only 3B active per token. the dense model is slower and it doesn't matter. watch 13 minutes of autonomous coding on a consumer GPU with zero intervention and tell me speed is what matters. a year ago this wasn't possible. now it runs on hardware you can buy used for $900. no API. no subscription. no cloud. just a 3090 doing what data centers did 18 months ago. full unedited session in the video. every token, every file, every thinking chain. 16 minutes. hit play.

Sudo su

91,135 Aufrufe • vor 5 Monaten

Chamath is making one of the most important business arguments of 2026. Half of large US companies right now cannot generate returns that exceed their cost of capital, which has normalized back to its long run average of 8 to 11%. Another one in seven companies globally is stuck generating persistent returns between 1 and 5% and most businesses don't have room for error and in this environment walks every frontier AI lab saying the same thing, give us your data, your workflows, your processes and our model will make everything better. And companies by the millions said yes. What they didn't fully account for is what happens on the other side of that door. Every time an employee runs a query through a frontier model API, the prompt goes through external servers, workflows, customer data, pricing logic, internal processes, all of it transmitted through a third party. As Alex Karp said companies are spending on tokens while handing over the exact proprietary advantages that make their business worth owning. Microsoft blocked internal use of Anthropic's Claude Fable 5 but over its 30-day data retention policy and the largest software company in the world decided a frontier model's data handling was too risky for its own employees. A US government action revoked access to another frontier model for foreign nationals overnight. Now here's where the cost math becomes impossible to ignore. Deutsche Bank calculated a roughly 65x cost gap between frontier models like Claude Fable 5 at ~$3.25 per task and open-source alternatives at ~$0.05. For 90% of everyday enterprise tasks, performance is comparable. Open-weight models now match closed frontier systems on core agent tasks at roughly one-tenth the cost, a high-volume deployment that costs $250/day on Claude runs at $12/day on an open-source equivalent. Chamath Palihapitiya tested this directly by running a standard enterprise code migration task through an orchestration layer wrapping an open-source model came in 16.4x cheaper than using a frontier model directly.
5:24

Chamath is making one of the most important business arguments of 2026. Half of large US companies right now cannot generate returns that exceed their cost of capital, which has normalized back to its long run average of 8 to 11%. Another one in seven companies globally is stuck generating persistent returns between 1 and 5% and most businesses don't have room for error and in this environment walks every frontier AI lab saying the same thing, give us your data, your workflows, your processes and our model will make everything better. And companies by the millions said yes. What they didn't fully account for is what happens on the other side of that door. Every time an employee runs a query through a frontier model API, the prompt goes through external servers, workflows, customer data, pricing logic, internal processes, all of it transmitted through a third party. As Alex Karp said companies are spending on tokens while handing over the exact proprietary advantages that make their business worth owning. Microsoft blocked internal use of Anthropic's Claude Fable 5 but over its 30-day data retention policy and the largest software company in the world decided a frontier model's data handling was too risky for its own employees. A US government action revoked access to another frontier model for foreign nationals overnight. Now here's where the cost math becomes impossible to ignore. Deutsche Bank calculated a roughly 65x cost gap between frontier models like Claude Fable 5 at ~$3.25 per task and open-source alternatives at ~$0.05. For 90% of everyday enterprise tasks, performance is comparable. Open-weight models now match closed frontier systems on core agent tasks at roughly one-tenth the cost, a high-volume deployment that costs $250/day on Claude runs at $12/day on an open-source equivalent. Chamath Palihapitiya tested this directly by running a standard enterprise code migration task through an orchestration layer wrapping an open-source model came in 16.4x cheaper than using a frontier model directly.

Milk Road AI

280,929 Aufrufe • vor 1 Monat

Qwopus on a single RTX 3090. Claude Opus 4.6 reasoning distilled into Qwen 3.5 27B dense, running through Claude's own coding agent (claude code). 29-35 tok/s with thinking mode on. the jinja bug that kills thinking on base Qwen doesn't carry over. harness and model matched. the base model would pause mid task on Claude Code. just stop generating. that's why i ran it through OpenCode, which handles stalled states automatically. this distilled version doesn't stall. it waits for tool outputs, reads them, selfcorrects when something breaks, and keeps going. i gave it a benchmark analysis task. went 9 minutes autonomous. wrote a README nobody asked for. zero steering. video is 5x speed but fully uncut. if you have a 3090, you can run this right now. free. no API. no subscription. opus structured reasoning on localhost. octopus invaders is next. same prompt that base qwen passed in 13 minutes and hermes 4.3 failed on 2x the hardware. i want to see if the distillation changes the outcome or just the style. more data soon.
1:53

Qwopus on a single RTX 3090. Claude Opus 4.6 reasoning distilled into Qwen 3.5 27B dense, running through Claude's own coding agent (claude code). 29-35 tok/s with thinking mode on. the jinja bug that kills thinking on base Qwen doesn't carry over. harness and model matched. the base model would pause mid task on Claude Code. just stop generating. that's why i ran it through OpenCode, which handles stalled states automatically. this distilled version doesn't stall. it waits for tool outputs, reads them, selfcorrects when something breaks, and keeps going. i gave it a benchmark analysis task. went 9 minutes autonomous. wrote a README nobody asked for. zero steering. video is 5x speed but fully uncut. if you have a 3090, you can run this right now. free. no API. no subscription. opus structured reasoning on localhost. octopus invaders is next. same prompt that base qwen passed in 13 minutes and hermes 4.3 failed on 2x the hardware. i want to see if the distillation changes the outcome or just the style. more data soon.

Sudo su

295,349 Aufrufe • vor 5 Monaten

🚨 I DON'T KNOW WHAT PEOPLE ARE WAITING FOR! This guy just deployed a full GPU compute farm in a data center Not renting cloud instances, not paying per-token, just racks of consumer GPUs on warehouse shelving doing AI inference 24/7 The setup is wild: open-frame rigs with rows of white fans, PCIe riser cables, server-grade PSUs, and a PDU for power distribution. All mounted on red industrial shelving with raised-floor cooling underneath. It looks like a crypto mining op, because it probably was one Here's what's actually happening: > Ex-mining rigs are being redeployed for AI inference at scale > Each rig runs 6-8 GPUs on PCIe risers connected to a single motherboard > Fans pull cold air from the raised floor through the open frames > A PDU on each rack handles power distribution across all rigs > The whole thing runs in a colocation facility with cheap power rates The economics make sense: cloud GPU pricing runs $2-4/hr per card. Own the hardware, colocate it for $0.08-0.12/kWh, and you're printing compute at a fraction of the cost The crypto-to-AI pipeline is real. Same hardware, same facilities, same power contracts, completely different revenue model Bookmark this!
0:53

🚨 I DON'T KNOW WHAT PEOPLE ARE WAITING FOR! This guy just deployed a full GPU compute farm in a data center Not renting cloud instances, not paying per-token, just racks of consumer GPUs on warehouse shelving doing AI inference 24/7 The setup is wild: open-frame rigs with rows of white fans, PCIe riser cables, server-grade PSUs, and a PDU for power distribution. All mounted on red industrial shelving with raised-floor cooling underneath. It looks like a crypto mining op, because it probably was one Here's what's actually happening: > Ex-mining rigs are being redeployed for AI inference at scale > Each rig runs 6-8 GPUs on PCIe risers connected to a single motherboard > Fans pull cold air from the raised floor through the open frames > A PDU on each rack handles power distribution across all rigs > The whole thing runs in a colocation facility with cheap power rates The economics make sense: cloud GPU pricing runs $2-4/hr per card. Own the hardware, colocate it for $0.08-0.12/kWh, and you're printing compute at a fraction of the cost The crypto-to-AI pipeline is real. Same hardware, same facilities, same power contracts, completely different revenue model Bookmark this!

cristal

16,305 Aufrufe • vor 2 Monaten

I designed a new test specifically for multimodal models: fill out a paper form. And it's much harder than it sounds. This isn't typing into an electronic field that captures your text. The form is just an image. The model has to place each form element: text, checkmarks — at the correct pixel position on the canvas itself. Results: 🟢 Kimi K2.6 → done in 3:45, 16.7k output tokens 🟡 Step 3.7 Flash → half the fields, 57k output tokens 🔴 Gemini 3.5 Flash → 489k output tokens, never finished. I had to kill it. Gemini burned ~29x more output tokens than Kimi on the exact same task, and Kimi's was the only form that actually looked filled out. The test, a mocked application form, contains some challenging parts, such as one-character-per-box fields. I provided every model the same set of tools: > get canvas size > drop probe markers to find coordinates > add text > add checkmarks > move elements > take a screenshot anytime to check their own work > ... etc So it's vision + spatial reasoning + tool use + long context, all at once. Small models (Qwen, Gemma) can't really complete this test, so I skipped them. What happened: > Kimi nailed name, DOB, ID, gender, marital status, nationality, email, phone, address, postal code — placement slightly loose, but content correct. 15 turns. Clean. > Step got maybe half right — fields dropped, "United States" landed in the email line, data floating outside boxes. Burned 1.24M input tokens doing it (81 turns of re-reading the canvas). > Gemini almost got there visually... then spiraled. By turn 40 it was issuing a delete_elements call wiping element IDs 365–425, basically erasing its own work. 31 minutes, 489k output tokens, still streaming. Terminated. The takeaway isn't "Gemini bad." This test is indeed difficult. But token efficiency is capability now. A model that needs 30x the tokens and still can't converge is going to be 30x the cost in production. Kimi K2.6 just quietly did the thing.
0:38

I designed a new test specifically for multimodal models: fill out a paper form. And it's much harder than it sounds. This isn't typing into an electronic field that captures your text. The form is just an image. The model has to place each form element: text, checkmarks — at the correct pixel position on the canvas itself. Results: 🟢 Kimi K2.6 → done in 3:45, 16.7k output tokens 🟡 Step 3.7 Flash → half the fields, 57k output tokens 🔴 Gemini 3.5 Flash → 489k output tokens, never finished. I had to kill it. Gemini burned ~29x more output tokens than Kimi on the exact same task, and Kimi's was the only form that actually looked filled out. The test, a mocked application form, contains some challenging parts, such as one-character-per-box fields. I provided every model the same set of tools: > get canvas size > drop probe markers to find coordinates > add text > add checkmarks > move elements > take a screenshot anytime to check their own work > ... etc So it's vision + spatial reasoning + tool use + long context, all at once. Small models (Qwen, Gemma) can't really complete this test, so I skipped them. What happened: > Kimi nailed name, DOB, ID, gender, marital status, nationality, email, phone, address, postal code — placement slightly loose, but content correct. 15 turns. Clean. > Step got maybe half right — fields dropped, "United States" landed in the email line, data floating outside boxes. Burned 1.24M input tokens doing it (81 turns of re-reading the canvas). > Gemini almost got there visually... then spiraled. By turn 40 it was issuing a delete_elements call wiping element IDs 365–425, basically erasing its own work. 31 minutes, 489k output tokens, still streaming. Terminated. The takeaway isn't "Gemini bad." This test is indeed difficult. But token efficiency is capability now. A model that needs 30x the tokens and still can't converge is going to be 30x the cost in production. Kimi K2.6 just quietly did the thing.

stevibe

25,455 Aufrufe • vor 3 Monaten

We made a thing! Very happy to announce sqlcoder-pro and the Defog Alignment Platform. Available to use immediately without a wait-list, weights will be open-sourced very soon. The video does a quick show and tell comparison against ChatGPT (with gpt-4o). Read on for more details! TLDR 💪 equal (or better) performance on text-to-SQL as the most capable Claude-3.5 or GPT-4 models 🤝 You can use it today on a free plan/free trial, without a waitlist 🪽 self-hostable on a single RTX4090, with 2 second median generation times for SQL queries 🔁 exactly the same output every time, give the same prompt 👨🏻‍🏫 teachable and steerable: show the model what you want it to do 🛞 debuggable – you can understand WTF is going on inside the model, instead of treating it like a black box Let's dig into each of these one-by-one! Performance SQLCoder-8b-pro significantly exceeds the performance of our previous sqlcoder-8b model on Postgres text-to-SQL (from 88.2% to 90.2% accuracy - gpt-4o is at 87.6%, for reference). It is also better at following instructions. This was done via self-merges, hand crafted fine-tuning data, and adapting the training data to fit our tokenizer. Cost You can host this on the model on a single $3,500 RTX4090, and support ~5 requests/second via VLLM. If you're looking to host on the cloud instead, you can run it on a single L4 GPU that costs $300/mo on GCP Repeatability We have a dense 8b model with no MoE shenanigans. For the same prompt with temperature=0, you'll always get the same answer – which is critical in BI. Teachable In our alignment and feedback modes, you can give the model feedback on how it answered certain questions, and it will automatically adapt to the feedback. Debuggable You can use logprobs and attention scores to determine where, exactly is the model paying attention to inside a prompt + what it's getting confused by when generating outputs. Available today You can use Defog on the cloud today by going to docs[dot]defog[dot]ai, and getting an API key. Excited to hear what you think!
2:33

We made a thing! Very happy to announce sqlcoder-pro and the Defog Alignment Platform. Available to use immediately without a wait-list, weights will be open-sourced very soon. The video does a quick show and tell comparison against ChatGPT (with gpt-4o). Read on for more details! TLDR 💪 equal (or better) performance on text-to-SQL as the most capable Claude-3.5 or GPT-4 models 🤝 You can use it today on a free plan/free trial, without a waitlist 🪽 self-hostable on a single RTX4090, with 2 second median generation times for SQL queries 🔁 exactly the same output every time, give the same prompt 👨🏻‍🏫 teachable and steerable: show the model what you want it to do 🛞 debuggable – you can understand WTF is going on inside the model, instead of treating it like a black box Let's dig into each of these one-by-one! Performance SQLCoder-8b-pro significantly exceeds the performance of our previous sqlcoder-8b model on Postgres text-to-SQL (from 88.2% to 90.2% accuracy - gpt-4o is at 87.6%, for reference). It is also better at following instructions. This was done via self-merges, hand crafted fine-tuning data, and adapting the training data to fit our tokenizer. Cost You can host this on the model on a single $3,500 RTX4090, and support ~5 requests/second via VLLM. If you're looking to host on the cloud instead, you can run it on a single L4 GPU that costs $300/mo on GCP Repeatability We have a dense 8b model with no MoE shenanigans. For the same prompt with temperature=0, you'll always get the same answer – which is critical in BI. Teachable In our alignment and feedback modes, you can give the model feedback on how it answered certain questions, and it will automatically adapt to the feedback. Debuggable You can use logprobs and attention scores to determine where, exactly is the model paying attention to inside a prompt + what it's getting confused by when generating outputs. Available today You can use Defog on the cloud today by going to docs[dot]defog[dot]ai, and getting an API key. Excited to hear what you think!

Rishabh Srivastava

13,465 Aufrufe • vor 2 Jahren

THIS GUY AUDITED 926 CLAUDE CODE SESSIONS AND FOUND MOST OF THE TOKEN WASTE WAS ON HIS SIDE everyone is blaming anthropic for the limits, so he decided to actually look at the data 858 sessions, 18,903 turns, and $1,619 estimated spend across 33 days here's what he found: 1\ one default setting was burning 14,000 tokens per turn Claude Code loads the full JSON schema for every tool into context at session start. whether you use them or not. 20,000 tokens of tool definitions sitting there on every single turn. the fix: one line in your settings.json "ENABLE_TOOL_SEARCH": "true" context dropped from 45K to 20K instantly. across 858 sessions that one setting was wasting an estimated 264 million tokens 2\ cache expiry is the single biggest waste 54% of his turns came after a 5+ minute idle gap. every one of those turns re-processed the entire conversation at full price which caused a 10x cost jump you go grab coffee. come back 5 minutes later. type your next message. everything rebuilds from scratch. the context didn't change. you didn't change. the cache just expired. 12.3 million tokens wasted on idle gaps alone 3\ 42 skills loaded. 19 of them used twice or less across 858 sessions. every one of those skill schemas sat in context on every turn eating tokens for nothing. 4\ 1,122 redundant file reads where the same file was read 3+ times one session read the same file 33 times. he ALSO built a full token auditor dashboard that shows you exactly where your waste is coming from 19 charts, opens in your browser, free AND open source
0:20

THIS GUY AUDITED 926 CLAUDE CODE SESSIONS AND FOUND MOST OF THE TOKEN WASTE WAS ON HIS SIDE everyone is blaming anthropic for the limits, so he decided to actually look at the data 858 sessions, 18,903 turns, and $1,619 estimated spend across 33 days here's what he found: 1\ one default setting was burning 14,000 tokens per turn Claude Code loads the full JSON schema for every tool into context at session start. whether you use them or not. 20,000 tokens of tool definitions sitting there on every single turn. the fix: one line in your settings.json "ENABLE_TOOL_SEARCH": "true" context dropped from 45K to 20K instantly. across 858 sessions that one setting was wasting an estimated 264 million tokens 2\ cache expiry is the single biggest waste 54% of his turns came after a 5+ minute idle gap. every one of those turns re-processed the entire conversation at full price which caused a 10x cost jump you go grab coffee. come back 5 minutes later. type your next message. everything rebuilds from scratch. the context didn't change. you didn't change. the cache just expired. 12.3 million tokens wasted on idle gaps alone 3\ 42 skills loaded. 19 of them used twice or less across 858 sessions. every one of those skill schemas sat in context on every turn eating tokens for nothing. 4\ 1,122 redundant file reads where the same file was read 3+ times one session read the same file 33 times. he ALSO built a full token auditor dashboard that shows you exactly where your waste is coming from 19 charts, opens in your browser, free AND open source

Om Patel

299,832 Aufrufe • vor 4 Monaten

Matthew Gallagher Built a $401M Company in Year One with 2 People. And the tool behind it? Claude Code. This year he's on track for $1.8B. Sam Altman predicted this. It's happening now. The problem? It costs money. API credits stack up. Monthly bills keep growing. Every prompt eats your budget. Every project drains your wallet faster. Until now. Two methods. 99% cheaper. One is completely free. Forever. $0. Not a trial. This video breaks down both step by step. ↓ Let me put this in perspective. $100-$500. That's monthly. That's what you spend. That's $6,000/year on API credits. Just to use a tool you haven't shipped anything with. The $401M guy? Spending $0. Same capability. Shipping weekly. Different cost structure. Different results. Different life. I'm about to hand you his cost structure for free. ↓ Open source vs closed source. Pay attention. Closed source: Claude. GPT-4. Pay per token. Meter always running. Open source: Qwen. Llama. Mistral. Free to download. Free to run. Free forever. No meter. No tokens. No bill. Here's what nobody tells you: 80% of coding tasks? Open source handles them. More than handles them. Writes clean code. Debugs errors. Generates boilerplate. Handles routine work perfectly. You're paying premium prices for tasks that don't need premium intelligence. That's hiring a brain surgeon to put on a bandaid. Smart play: Free models for the 80%. Paid credits for the 20%. That's what the $401M guy does. That's what this video teaches you. Follow Himanshu Kumar for more breakdowns that turn free tools into real businesses. ↓ Method 1: Ollama. Local. Free. Forever. Download it. Pull a model. Point Claude Code at it. Done. No internet needed. No API keys required. No monthly subscription. No token counting ever. No bill. Today. Tomorrow. Ever. Your data never leaves your computer. Complete privacy. Complete freedom. Claude Code thinks it's talking to the cloud. It's talking to your laptop. For $0. The video walks through every step: Every config file. Every variable. Every command. Every click. If you can follow a recipe, you can do this. People who set this up 3 months ago? Saved $300-$1,500 since then. Workflow didn't change one bit. ↓ Hardware you need: 16GB RAM: 7B models run smooth. 32GB RAM: 32B models run comfortable. 64GB + GPU: biggest models available. No GPU? Still works. Just slower. Few extra seconds. That's it. Your $1,500 laptop is sitting there running Chrome and Spotify. Put it to work saving you $200/month instead. Follow Himanshu Kumar for more breakdowns that turn free tools into real businesses. ↓ Method 2: Open Router. Free Cloud. No Hardware. Weak machine? Don't want local setup? This method is for you. Free AI models in the cloud. No download. No hardware. Configure Claude Code to route through Open Router. The config: Base URL: Open Router API. API key: free Open Router key. Default Sonnet: free. Default Opus: free. Default Haiku: free. Small fast model: free. Subagent model: free. Free. Free. Free. Free. Free across the board. Same interface. Same commands. Same workflow. Zero cost. Copy the config from the video. Paste it. Save $200/month. Starting today. Right now. ↓ When to use which: Ollama (local): Best for privacy. Best for offline work. Best for unlimited usage. Best if you have decent hardware. Open Router (cloud): Best for weak machines. Best for instant setup. Best for trying different models. Best if you don't want to manage anything. Both methods: Best for 80% of your daily work. Still use paid Claude for: Complex architecture. Multi-file refactoring. Deep reasoning tasks. The 20% that actually needs it. $20/month instead of $200/month. Same output. 90% less cost. ↓ The math that should make you angry. You (current): $200-$500/month. $2,400-$6,000/year. $7,200-$18,000 over 3 years. You (after this video): $20-$50/month. $240-$600/year. $720-$1,800 over 3 years. Savings over 3 years: $6,480-$16,200. That's a used car. That's seed money. That's 6 months of rent. All from one 25-minute video. All from 15 minutes of configuration. Highest ROI 25 minutes you'll spend this year. ↓ The limitations. I won't lie to you. Open source is not Opus. Not as smart on complex reasoning. Not as good at long-context tasks. Makes more mistakes on nuanced problems. But they are: Free. Capable. Getting better monthly. Good enough for 80% of daily work. Smart cost management isn't being cheap. It's being strategic. Expensive tool when it matters. Free tool when it doesn't. ↓ The one-person billion-dollar company is coming. $401M in year one proved it's possible. The building blocks: AI that codes: Claude Code. Way to run it free: this video. Distribution: the internet. Customers: everyone. Only missing ingredient? Someone who builds. Not reads about building. Not saves posts about building. Not bookmarks videos about building. Builds. Tools are free. Knowledge is free. Opportunity is screaming. You're still "thinking about it." ↓ Your action plan: Tonight: Watch the video. Tomorrow morning: Set up Ollama or Open Router. Tomorrow afternoon: Build something. Anything. This week: Build a second thing. Faster. This month: Charge someone for it. One video. One setup. One weekend. $0 cost. Unlimited potential. Or keep paying $200/month for something you could get free. Keep consuming instead of building. Keep planning instead of shipping. Matthew Gallagher didn't plan a $401M company. He built it. Full video attached. Every method. Every config. Every tradeoff. 25 minutes. Your move. Follow Himanshu Kumar for more breakdowns that turn free tools into real businesses.
25:22

Matthew Gallagher Built a $401M Company in Year One with 2 People. And the tool behind it? Claude Code. This year he's on track for $1.8B. Sam Altman predicted this. It's happening now. The problem? It costs money. API credits stack up. Monthly bills keep growing. Every prompt eats your budget. Every project drains your wallet faster. Until now. Two methods. 99% cheaper. One is completely free. Forever. $0. Not a trial. This video breaks down both step by step. ↓ Let me put this in perspective. $100-$500. That's monthly. That's what you spend. That's $6,000/year on API credits. Just to use a tool you haven't shipped anything with. The $401M guy? Spending $0. Same capability. Shipping weekly. Different cost structure. Different results. Different life. I'm about to hand you his cost structure for free. ↓ Open source vs closed source. Pay attention. Closed source: Claude. GPT-4. Pay per token. Meter always running. Open source: Qwen. Llama. Mistral. Free to download. Free to run. Free forever. No meter. No tokens. No bill. Here's what nobody tells you: 80% of coding tasks? Open source handles them. More than handles them. Writes clean code. Debugs errors. Generates boilerplate. Handles routine work perfectly. You're paying premium prices for tasks that don't need premium intelligence. That's hiring a brain surgeon to put on a bandaid. Smart play: Free models for the 80%. Paid credits for the 20%. That's what the $401M guy does. That's what this video teaches you. Follow Himanshu Kumar for more breakdowns that turn free tools into real businesses. ↓ Method 1: Ollama. Local. Free. Forever. Download it. Pull a model. Point Claude Code at it. Done. No internet needed. No API keys required. No monthly subscription. No token counting ever. No bill. Today. Tomorrow. Ever. Your data never leaves your computer. Complete privacy. Complete freedom. Claude Code thinks it's talking to the cloud. It's talking to your laptop. For $0. The video walks through every step: Every config file. Every variable. Every command. Every click. If you can follow a recipe, you can do this. People who set this up 3 months ago? Saved $300-$1,500 since then. Workflow didn't change one bit. ↓ Hardware you need: 16GB RAM: 7B models run smooth. 32GB RAM: 32B models run comfortable. 64GB + GPU: biggest models available. No GPU? Still works. Just slower. Few extra seconds. That's it. Your $1,500 laptop is sitting there running Chrome and Spotify. Put it to work saving you $200/month instead. Follow Himanshu Kumar for more breakdowns that turn free tools into real businesses. ↓ Method 2: Open Router. Free Cloud. No Hardware. Weak machine? Don't want local setup? This method is for you. Free AI models in the cloud. No download. No hardware. Configure Claude Code to route through Open Router. The config: Base URL: Open Router API. API key: free Open Router key. Default Sonnet: free. Default Opus: free. Default Haiku: free. Small fast model: free. Subagent model: free. Free. Free. Free. Free. Free across the board. Same interface. Same commands. Same workflow. Zero cost. Copy the config from the video. Paste it. Save $200/month. Starting today. Right now. ↓ When to use which: Ollama (local): Best for privacy. Best for offline work. Best for unlimited usage. Best if you have decent hardware. Open Router (cloud): Best for weak machines. Best for instant setup. Best for trying different models. Best if you don't want to manage anything. Both methods: Best for 80% of your daily work. Still use paid Claude for: Complex architecture. Multi-file refactoring. Deep reasoning tasks. The 20% that actually needs it. $20/month instead of $200/month. Same output. 90% less cost. ↓ The math that should make you angry. You (current): $200-$500/month. $2,400-$6,000/year. $7,200-$18,000 over 3 years. You (after this video): $20-$50/month. $240-$600/year. $720-$1,800 over 3 years. Savings over 3 years: $6,480-$16,200. That's a used car. That's seed money. That's 6 months of rent. All from one 25-minute video. All from 15 minutes of configuration. Highest ROI 25 minutes you'll spend this year. ↓ The limitations. I won't lie to you. Open source is not Opus. Not as smart on complex reasoning. Not as good at long-context tasks. Makes more mistakes on nuanced problems. But they are: Free. Capable. Getting better monthly. Good enough for 80% of daily work. Smart cost management isn't being cheap. It's being strategic. Expensive tool when it matters. Free tool when it doesn't. ↓ The one-person billion-dollar company is coming. $401M in year one proved it's possible. The building blocks: AI that codes: Claude Code. Way to run it free: this video. Distribution: the internet. Customers: everyone. Only missing ingredient? Someone who builds. Not reads about building. Not saves posts about building. Not bookmarks videos about building. Builds. Tools are free. Knowledge is free. Opportunity is screaming. You're still "thinking about it." ↓ Your action plan: Tonight: Watch the video. Tomorrow morning: Set up Ollama or Open Router. Tomorrow afternoon: Build something. Anything. This week: Build a second thing. Faster. This month: Charge someone for it. One video. One setup. One weekend. $0 cost. Unlimited potential. Or keep paying $200/month for something you could get free. Keep consuming instead of building. Keep planning instead of shipping. Matthew Gallagher didn't plan a $401M company. He built it. Full video attached. Every method. Every config. Every tradeoff. 25 minutes. Your move. Follow Himanshu Kumar for more breakdowns that turn free tools into real businesses.

Himanshu Kumar

13,677 Aufrufe • vor 4 Monaten

Claude Tag has completely changed the way I do work for the last 4 months. Except… it's not Claude Tag. Anthropic only announced that a few hours ago, and I don't even have access yet. But I did build a version of it for myself which I've been using for months now. Here's how. 4 months ago, inspired by the success of OpenClaw, I wondered what would happen if I let Claude Code on its own computer 24x7. So I built a simple harness that allowed me to turn any Mac into an AI employee with Claude Code headless mode (-p). Today, I manage 3 such AI employees. It started with Luo Ji — my and my brother Piyush Agarwal's AI co-founder, running in our personal Slack. Luo does real work for us. We've been writing a 100% of the code for 3 products on Slack with Luo now. It manages our emails and gives us a little brief each day with things we need to take action on. And so much more. And it's not just the two of us. On the consulting team at Every 🪨, we run Claudie and for the editorial team, Andy. Same architecture, same Slack, months of real work. They help the teams with work related to project management, chief-of-staff work, data hygiene, building decks, writing first drafts, even browsing X on their own account for AI updates. So it's mindblowing to see that Anthropic landed on the exact same architecture I did. Claude Tag is an AI employee that lives in your Slack workspace and does work autonomously. Anthropic says they've been running it internally for the better part of this year — opening PRs, doing real work. And so have I. So has my whole team. The architectural decisions Anthropic baked into Claude Tag are the ones we arrived at too: - Built on Claude Code - Uses its own accounts - A separate employee per team - Slack as the interface This is the future of work, and I've been living it for months. I've shifted all of my workflows — code, PRs, even the non-technical stuff — out of Claude Code and the Claude app and into Slack. I've had entire weeks where I never opened Claude Code on my laptop. Here's a video walkthrough of how I've been using this in real life.
14:36

Claude Tag has completely changed the way I do work for the last 4 months. Except… it's not Claude Tag. Anthropic only announced that a few hours ago, and I don't even have access yet. But I did build a version of it for myself which I've been using for months now. Here's how. 4 months ago, inspired by the success of OpenClaw, I wondered what would happen if I let Claude Code on its own computer 24x7. So I built a simple harness that allowed me to turn any Mac into an AI employee with Claude Code headless mode (-p). Today, I manage 3 such AI employees. It started with Luo Ji — my and my brother Piyush Agarwal's AI co-founder, running in our personal Slack. Luo does real work for us. We've been writing a 100% of the code for 3 products on Slack with Luo now. It manages our emails and gives us a little brief each day with things we need to take action on. And so much more. And it's not just the two of us. On the consulting team at Every 🪨, we run Claudie and for the editorial team, Andy. Same architecture, same Slack, months of real work. They help the teams with work related to project management, chief-of-staff work, data hygiene, building decks, writing first drafts, even browsing X on their own account for AI updates. So it's mindblowing to see that Anthropic landed on the exact same architecture I did. Claude Tag is an AI employee that lives in your Slack workspace and does work autonomously. Anthropic says they've been running it internally for the better part of this year — opening PRs, doing real work. And so have I. So has my whole team. The architectural decisions Anthropic baked into Claude Tag are the ones we arrived at too: - Built on Claude Code - Uses its own accounts - A separate employee per team - Slack as the interface This is the future of work, and I've been living it for months. I've shifted all of my workflows — code, PRs, even the non-technical stuff — out of Claude Code and the Claude app and into Slack. I've had entire weeks where I never opened Claude Code on my laptop. Here's a video walkthrough of how I've been using this in real life.

Nityesh

36,790 Aufrufe • vor 2 Monaten

Chatgpt Vs Claude in Excel TLDR: Gave ChatGPT and Claude the same credit risk modeling task in Excel. ChatGPT followed proper methodology, did variable clustering, train/test split, Excel native model, AUC 0.627 on test data. Claude skipped clustering, inflated IV with sparse categories, ran the model in JavaScript not Excel, reported 0.706 AUC on training data with no split. Chagpt 5.4 nailed a real world Banking and Finance modelling case study. I gave both ChatGPT and Claude the same banking credit risk dataset with the same prompt: bin the variables, calculate Information Value, do variable clustering, select features, build a model in Excel, generate ROC. Basically an end to end scorecard development workflow inside Excel. ChatGPT followed the brief properly. Quantile-based WoE binning, IV ranking across all 33 variables, correlation-based variable clustering at 0.75 threshold to remove redundant features, picked 6 representative variables, built a 2-variable Excel-native decision tree with proper train/test split via ID mod 10, scored leaf-level bad rates, ROC on held-out test data. AUC: 0.627. Claude looked more impressive on the surface. 28 sheets, individual binning for every variable, logistic regression with 27 features, gradient descent at 500 epochs, full coefficient table with importance bars, confusion matrix, precision/recall/F1. AUC: 0.7066. Sounds like it won right? So, The logistic regression was computed in JavaScript via execute_office_js and results were pasted as static values. That's not Excel-native, against the task. The 0.706 AUC? On the full training set, no train/test split. That number is meaningless. Variable clustering? Skipped entirely even though the prompt explicitly asked for it. It "selected" 31 out of 33 features which is barely feature selection. The entire feature ranking was built on a methodological error. I've done this exact dataset myself with deeper feature engineering and ensemble technique hit 0.70 AUC. The fact that ChatGPT got to 0.627 with less feature enigineering and naive modelling is genuinely impressive. Binning and IV based feature selection is one of the most important techniques in credit risk modeling and it nailed the workflow. One thing claude did well is the workbook is visually impressive. One area ChatGPT can improve, the visuals and formatting could be sharper, the workbook is functional but not polished.
1:51

Chatgpt Vs Claude in Excel TLDR: Gave ChatGPT and Claude the same credit risk modeling task in Excel. ChatGPT followed proper methodology, did variable clustering, train/test split, Excel native model, AUC 0.627 on test data. Claude skipped clustering, inflated IV with sparse categories, ran the model in JavaScript not Excel, reported 0.706 AUC on training data with no split. Chagpt 5.4 nailed a real world Banking and Finance modelling case study. I gave both ChatGPT and Claude the same banking credit risk dataset with the same prompt: bin the variables, calculate Information Value, do variable clustering, select features, build a model in Excel, generate ROC. Basically an end to end scorecard development workflow inside Excel. ChatGPT followed the brief properly. Quantile-based WoE binning, IV ranking across all 33 variables, correlation-based variable clustering at 0.75 threshold to remove redundant features, picked 6 representative variables, built a 2-variable Excel-native decision tree with proper train/test split via ID mod 10, scored leaf-level bad rates, ROC on held-out test data. AUC: 0.627. Claude looked more impressive on the surface. 28 sheets, individual binning for every variable, logistic regression with 27 features, gradient descent at 500 epochs, full coefficient table with importance bars, confusion matrix, precision/recall/F1. AUC: 0.7066. Sounds like it won right? So, The logistic regression was computed in JavaScript via execute_office_js and results were pasted as static values. That's not Excel-native, against the task. The 0.706 AUC? On the full training set, no train/test split. That number is meaningless. Variable clustering? Skipped entirely even though the prompt explicitly asked for it. It "selected" 31 out of 33 features which is barely feature selection. The entire feature ranking was built on a methodological error. I've done this exact dataset myself with deeper feature engineering and ensemble technique hit 0.70 AUC. The fact that ChatGPT got to 0.627 with less feature enigineering and naive modelling is genuinely impressive. Binning and IV based feature selection is one of the most important techniques in credit risk modeling and it nailed the workflow. One thing claude did well is the workbook is visually impressive. One area ChatGPT can improve, the visuals and formatting could be sharper, the workbook is functional but not polished.

Akshen

35,174 Aufrufe • vor 5 Monaten

Here's GPT-5.6 Sol vs Fable asked to build the exact same app (prompts in the article!) What do you think? I have my opinions, but I'd like to hear yours. Both built complete, working apps, but with very different approaches: - Sol took less than half the time at the same (high) reasoning - Sol spent much less time on frame analysis. Fable dumped sheets of frames for every transition and scrutinized them multiple times; Sol only checked twice, and both times was just like "looks great, continuing" - At first, Sol intentionally ignored my request to use image generation! It went with procedurally generated food images "for simplicity" (they looked terrible) - On the second prompt, Sol generated images, but intentionally ignored my suggestion for transparent image backgrounds, saying full photos "preserve shadows more convincingly" - Sol's used its built-in GPT Image 2 integration, which saved cost; Fable had to use the Gemini API - Sol didn't seem to know how to use Liquid Glass at all - Fable was more restrained; Sol wanted to add extra labels and detail - On the other hand, Fable decided to add an extra detail page for foods, while Sol did not - Both had nice haptics on device - Fable added a pinch dismiss gesture; Sol did not do gestures
0:31

Here's GPT-5.6 Sol vs Fable asked to build the exact same app (prompts in the article!) What do you think? I have my opinions, but I'd like to hear yours. Both built complete, working apps, but with very different approaches: - Sol took less than half the time at the same (high) reasoning - Sol spent much less time on frame analysis. Fable dumped sheets of frames for every transition and scrutinized them multiple times; Sol only checked twice, and both times was just like "looks great, continuing" - At first, Sol intentionally ignored my request to use image generation! It went with procedurally generated food images "for simplicity" (they looked terrible) - On the second prompt, Sol generated images, but intentionally ignored my suggestion for transparent image backgrounds, saying full photos "preserve shadows more convincingly" - Sol's used its built-in GPT Image 2 integration, which saved cost; Fable had to use the Gemini API - Sol didn't seem to know how to use Liquid Glass at all - Fable was more restrained; Sol wanted to add extra labels and detail - On the other hand, Fable decided to add an extra detail page for foods, while Sol did not - Both had nice haptics on device - Fable added a pinch dismiss gesture; Sol did not do gestures

Anshu

161,283 Aufrufe • vor 1 Monat

Chinese AI models are wiping billions off Big Tech right now. Google just lost $200 billion in a single day, and the model it needed to fight back still isn't ready. Gemini 3.5 Pro, Google's most powerful model, is months behind schedule. Alphabet stock dropped 4.4% that same day. The Deepseek moment is happening again, and the new model is FAR bigger. On the same day Google's delay leaked, a Beijing lab called Moonshot released Kimi K3. It is the largest open model ever built, with 2.8 trillion parameters. It took the number one spot on the Frontend Code Arena, a live coding leaderboard, passing Anthropic's best model. And Moonshot is giving it away for free on July 27. The genius part: Anyone with enough computers can download it and run a frontier level AI without paying a cent to a US company. A single task on Kimi K3 costs about 94 cents. The same work on some American models costs nearly double. So why would a company keep paying premium prices for a model it can now get for free? The entire US AI business is built on selling access to models that cost billions to train. If a free Chinese version does most of the same work, that pricing power starts to crack. And Kimi is close to the best. On one closely watched intelligence ranking it scored 57, just behind the top American models GPT-5.6 Sol and Fable 5, and ahead of Claude Opus 4.8. Bank of America told clients that Kimi proves Chinese labs can keep making big leaps even with limited chips. And the founder of Moonshot, Yang Zhilin, learned to build AI as a researcher INSIDE Google. Google literally wrote the 2017 paper that made all of these models possible. Now the people who studied its work are using it to destroy Google, and handing it out for free. What happens next: Kimi K3's weights go public on July 27. Google reports earnings on July 22, and everyone will be asking the same question about Gemini. If free models keep topping the charts, every valuation built on paid AI access has to be rewritten. What do you think?
1:10

Chinese AI models are wiping billions off Big Tech right now. Google just lost $200 billion in a single day, and the model it needed to fight back still isn't ready. Gemini 3.5 Pro, Google's most powerful model, is months behind schedule. Alphabet stock dropped 4.4% that same day. The Deepseek moment is happening again, and the new model is FAR bigger. On the same day Google's delay leaked, a Beijing lab called Moonshot released Kimi K3. It is the largest open model ever built, with 2.8 trillion parameters. It took the number one spot on the Frontend Code Arena, a live coding leaderboard, passing Anthropic's best model. And Moonshot is giving it away for free on July 27. The genius part: Anyone with enough computers can download it and run a frontier level AI without paying a cent to a US company. A single task on Kimi K3 costs about 94 cents. The same work on some American models costs nearly double. So why would a company keep paying premium prices for a model it can now get for free? The entire US AI business is built on selling access to models that cost billions to train. If a free Chinese version does most of the same work, that pricing power starts to crack. And Kimi is close to the best. On one closely watched intelligence ranking it scored 57, just behind the top American models GPT-5.6 Sol and Fable 5, and ahead of Claude Opus 4.8. Bank of America told clients that Kimi proves Chinese labs can keep making big leaps even with limited chips. And the founder of Moonshot, Yang Zhilin, learned to build AI as a researcher INSIDE Google. Google literally wrote the 2017 paper that made all of these models possible. Now the people who studied its work are using it to destroy Google, and handing it out for free. What happens next: Kimi K3's weights go public on July 27. Google reports earnings on July 22, and everyone will be asking the same question about Gemini. If free models keep topping the charts, every valuation built on paid AI access has to be rewritten. What do you think?

Ricardo

47,790 Aufrufe • vor 1 Monat

this is what 12 gigs of VRAM built in 2026. a 9 billion parameter model running on a 5 year old RTX 3060 wrote a full space shooter from a single prompt. blank screen on first try. i came back with a bug list and the same model on the same card fixed every issue across 11 files without touching a single line myself. enemies still looked wrong so i pushed another iteration and now the game has pixel art octopi, particle effects, screen shake, projectile physics and a combo system. all running locally on a card that was designed to play fortnite. three iterations. zero cloud. zero API calls. every token generated on hardware sitting under my desk. the model reads its own code, finds what's broken, patches it, validates syntax and restarts the server. i just describe what's wrong and it handles the rest. people are paying monthly subscriptions to type into a browser tab and wait for a server farm to respond. meanwhile a GPU you can find used on ebay is running a full autonomous hermes agent framework with 31 tools, 128K context window and thinking mode generating at 29 tokens per second nonstop. the game still needs work. level upgrades don't trigger and boss fights need tuning. but the fact that i'm iterating on gameplay balance instead of debugging whether the code runs at all tells you where this is headed. every iteration the game gets better on the same hardware. same 12 gigs. same 9 billion parameters. same RTX 3060 from 5 years ago your GPU is not a gaming card anymore. it's a local AI lab that never sends your data anywhere.
0:28

this is what 12 gigs of VRAM built in 2026. a 9 billion parameter model running on a 5 year old RTX 3060 wrote a full space shooter from a single prompt. blank screen on first try. i came back with a bug list and the same model on the same card fixed every issue across 11 files without touching a single line myself. enemies still looked wrong so i pushed another iteration and now the game has pixel art octopi, particle effects, screen shake, projectile physics and a combo system. all running locally on a card that was designed to play fortnite. three iterations. zero cloud. zero API calls. every token generated on hardware sitting under my desk. the model reads its own code, finds what's broken, patches it, validates syntax and restarts the server. i just describe what's wrong and it handles the rest. people are paying monthly subscriptions to type into a browser tab and wait for a server farm to respond. meanwhile a GPU you can find used on ebay is running a full autonomous hermes agent framework with 31 tools, 128K context window and thinking mode generating at 29 tokens per second nonstop. the game still needs work. level upgrades don't trigger and boss fights need tuning. but the fact that i'm iterating on gameplay balance instead of debugging whether the code runs at all tells you where this is headed. every iteration the game gets better on the same hardware. same 12 gigs. same 9 billion parameters. same RTX 3060 from 5 years ago your GPU is not a gaming card anymore. it's a local AI lab that never sends your data anywhere.

Sudo su

170,848 Aufrufe • vor 5 Monaten

ANTHROPIC LEAKED A FILE THEY SPENT 2 YEARS BUILDING WHERE 4 DEPARTMENTS CLOSE 90% OF YOUR WORK FOR $4 A DAY you do only 10% - the deciding, and the other 90% runs on different versions of Claude at a fraction of the price. research → marketing → sales → finance → back into the file research pulls in hundreds of sources and a handful survive - the cheap model filters, the expensive one reads only what got through. paying the top rate for a page you'll discard anyway is the commonest overspend there is. marketing is high volume with a low stake per item - so you generate 3 variants at once instead of one. one variant is a guess, three variants are a choice, and they cost the same. in sales the cheap model scores 400 leads and the expensive one writes only to the few worth it. a personal letter to a lead that was never going to convert is a paid guess. 6 of every 8 tasks in this file never need the priciest model - it answers at 5x the rate of the cheap one. and there's exactly one department where saving on the model makes no sense - unit economics, funnel, forecast. one wrong number here costs more than a full year of token savings. finance is the only department that writes back into the file - next month's marketing runs on rules analytics wrote, not you. overpaying is annoying, underpaying is expensive - route by the cost of being wrong, not the price per token. save this and paste it into Claude Code - 4 departments execute, you choose ↓
1:02

ANTHROPIC LEAKED A FILE THEY SPENT 2 YEARS BUILDING WHERE 4 DEPARTMENTS CLOSE 90% OF YOUR WORK FOR $4 A DAY you do only 10% - the deciding, and the other 90% runs on different versions of Claude at a fraction of the price. research → marketing → sales → finance → back into the file research pulls in hundreds of sources and a handful survive - the cheap model filters, the expensive one reads only what got through. paying the top rate for a page you'll discard anyway is the commonest overspend there is. marketing is high volume with a low stake per item - so you generate 3 variants at once instead of one. one variant is a guess, three variants are a choice, and they cost the same. in sales the cheap model scores 400 leads and the expensive one writes only to the few worth it. a personal letter to a lead that was never going to convert is a paid guess. 6 of every 8 tasks in this file never need the priciest model - it answers at 5x the rate of the cheap one. and there's exactly one department where saving on the model makes no sense - unit economics, funnel, forecast. one wrong number here costs more than a full year of token savings. finance is the only department that writes back into the file - next month's marketing runs on rules analytics wrote, not you. overpaying is annoying, underpaying is expensive - route by the cost of being wrong, not the price per token. save this and paste it into Claude Code - 4 departments execute, you choose ↓

Sprytix

27,027 Aufrufe • vor 13 Tagen

My AI made Shopify pages are getting better and better everyday Here’s an example of a Shopify section I built with the reference page and the result on my store 👇 Guide: To make good AI landing pages, you need to use the same method as making good AI UGC or AI product images you have to take something that's already good as a reference for Claude/gemini to analyze and adapt for your product/brand for now you can't adapt full landing pages to your product because the Claude context gets bloated fast and you get poor output. Listicles are the only kind of page you can "one shot" with Claude. But Product pages are another story Analyze an existing page and adapting it to your product section by section is the way to go. The results are 10x better. here's an example (on the video): 1- I found this good product section from im8's product page. I screen-recorded the section on both desktop and mobile, going through the animations to capture the dynamism that a screenshot wouldn’t show. Then I asked Claude Web or Gemini to analyze the recording and produce a very detailed report. Full prompt is on my TG channel, it's quite long. 2- then ask Claude (inside your Shopify brand project folder): "I have a detailed UI/UX specification document for a product page section I want to adapt to my product. Recreate this exactly on Shopify as a section template and adapt it to [name of the product] using the brand guidelines" (paste the result prompt from step 1 ). 3- you will have your section ready after 3-4 minutes. you'll probably have to change a few things. spacing, small visual bugs, price not appearing correctly. it will take you 5 minutes maximum. 4- then you can ask Claude to make 4 different variations of the section using different designs and pick the best one using this prompt: "Create 4 design variations of this section. Keep the content and layout structure identical across all, only vary the visual treatment (color usage, typography hierarchy, spacing, component styling). I will choose the one I like the most." after that you have a pretty good section, and you can do the process again for all sections of the page. The less complex the section, the faster the process will be. note: I know the AI result is not perfect, but it's pretty impressive imo and it will only get better.
0:27

My AI made Shopify pages are getting better and better everyday Here’s an example of a Shopify section I built with the reference page and the result on my store 👇 Guide: To make good AI landing pages, you need to use the same method as making good AI UGC or AI product images you have to take something that's already good as a reference for Claude/gemini to analyze and adapt for your product/brand for now you can't adapt full landing pages to your product because the Claude context gets bloated fast and you get poor output. Listicles are the only kind of page you can "one shot" with Claude. But Product pages are another story Analyze an existing page and adapting it to your product section by section is the way to go. The results are 10x better. here's an example (on the video): 1- I found this good product section from im8's product page. I screen-recorded the section on both desktop and mobile, going through the animations to capture the dynamism that a screenshot wouldn’t show. Then I asked Claude Web or Gemini to analyze the recording and produce a very detailed report. Full prompt is on my TG channel, it's quite long. 2- then ask Claude (inside your Shopify brand project folder): "I have a detailed UI/UX specification document for a product page section I want to adapt to my product. Recreate this exactly on Shopify as a section template and adapt it to [name of the product] using the brand guidelines" (paste the result prompt from step 1 ). 3- you will have your section ready after 3-4 minutes. you'll probably have to change a few things. spacing, small visual bugs, price not appearing correctly. it will take you 5 minutes maximum. 4- then you can ask Claude to make 4 different variations of the section using different designs and pick the best one using this prompt: "Create 4 design variations of this section. Keep the content and layout structure identical across all, only vary the visual treatment (color usage, typography hierarchy, spacing, component styling). I will choose the one I like the most." after that you have a pretty good section, and you can do the process again for all sections of the page. The less complex the section, the faster the process will be. note: I know the AI result is not perfect, but it's pretty impressive imo and it will only get better.

Olivier

84,679 Aufrufe • vor 5 Monaten

$47K Profit in 3 Months. My Bot Sees the Future 8 Seconds Before Polymarket Does. Penalty. Red card. Goal. Probability jumps 15-30 points instantly. Polymarket updates in 15-20 seconds. My bot updates in 8. That 8-second gap is free money. And you're still watching matches and placing bets manually like it's a pub quiz. I've made the exact step-by-step guide to build this live sports arbitrage bot. Free for 24 hours. To get it: 1. Comment "Polymarket" 2. Like and Retweet 3. Follow me Himanshu Kumar (so i can DM you) How it works. Live stadium data through Sportradar WebSocket. Not broadcast data. Not TV delay. Actual real-time player coordinates, speed, xG, pressing intensity. Same data bookmakers buy. $12K/year. ML model recalculates probability after every event. Lineup. Form. Fatigue. Minute. Score. 3 months of training. $2K on GPU. Private RPC on Polygon. $200/mo. 3-5 seconds faster than public nodes. Gas overpay for block priority. $800/mo at 8 matches per week. Up to 8 matches running in parallel. Champions League. EPL. La Liga. Each match a separate instance. Separate wallet. Total cost: $18K/year. But I see the future 8 seconds ahead. You're watching the same match. Seeing the same goal. By the time you react, my bot already closed the position and took the profit. I don't know who will win. But I know exactly where the oracle falls behind. That gap is where the money lives. And you can't even see it. You Must Follow me Himanshu Kumar, so i can send you DM.
0:36

$47K Profit in 3 Months. My Bot Sees the Future 8 Seconds Before Polymarket Does. Penalty. Red card. Goal. Probability jumps 15-30 points instantly. Polymarket updates in 15-20 seconds. My bot updates in 8. That 8-second gap is free money. And you're still watching matches and placing bets manually like it's a pub quiz. I've made the exact step-by-step guide to build this live sports arbitrage bot. Free for 24 hours. To get it: 1. Comment "Polymarket" 2. Like and Retweet 3. Follow me Himanshu Kumar (so i can DM you) How it works. Live stadium data through Sportradar WebSocket. Not broadcast data. Not TV delay. Actual real-time player coordinates, speed, xG, pressing intensity. Same data bookmakers buy. $12K/year. ML model recalculates probability after every event. Lineup. Form. Fatigue. Minute. Score. 3 months of training. $2K on GPU. Private RPC on Polygon. $200/mo. 3-5 seconds faster than public nodes. Gas overpay for block priority. $800/mo at 8 matches per week. Up to 8 matches running in parallel. Champions League. EPL. La Liga. Each match a separate instance. Separate wallet. Total cost: $18K/year. But I see the future 8 seconds ahead. You're watching the same match. Seeing the same goal. By the time you react, my bot already closed the position and took the profit. I don't know who will win. But I know exactly where the oracle falls behind. That gap is where the money lives. And you can't even see it. You Must Follow me Himanshu Kumar, so i can send you DM.

Himanshu Kumar

32,068 Aufrufe • vor 4 Monaten

Walmart's genius private label strategy literally KILLED name brands... By doing the exact OPPOSITE of what everyone said you should do. In April 2024, they launched Bettergoods - a "cheap" store brand. But instead of targeting broke customers, they went after RICH ones. 28% of U.S. households bought it in one year. $500 million in sales. 40% came back for more. It's now the fastest-growing private label brand in America. This is the most insane private label play in retail history... The Old Playbook (that everyone follows): - Private labels = cheaper version of name brands - Target price-conscious shoppers - Make it look "good enough" - Compete on cost alone Great Value, Kirkland, Equate - they all did this. And it worked. For the bottom 80% of customers. But Walmart saw something nobody else did. The Problem With Traditional Private Labels: High-income shoppers won't touch them. They associate store brands with "lower quality." Even when the product is IDENTICAL, the perception kills sales. So Walmart asked a different question: "What if we made a private label that high-income people actually WANT?" The Bettergoods Strategy: - Most items under $5 (same as Great Value) - But packaging looks like a $15 boutique product - "Chef-inspired" flavors - Trend-forward ingredients (oat milk ice cream, hot honey seasoning, plant-based everything) - Totally unique products - NOT knock-offs of name brands And here's the genius part: They're selling to TWO completely different customers with the SAME product. Low-income shoppers: "I'm treating myself without breaking the bank" High-income shoppers: "This is premium quality at an amazing price" The numbers prove it worked: High-income households are 20% MORE likely to buy Bettergoods than Great Value. Low-income households are 34% more likely to buy Bettergoods than other premium brands. They captured BOTH ends of the market. With the same $3.44 pint of ice cream. Why This Killed Name Brands: Before Bettergoods, high-income Walmart shoppers bought Ben & Jerry's, Häagen-Dazs, Talenti. Now they buy Bettergoods. Same quality. Half the price. Trendier flavors. The name brands lost their premium positioning overnight. Because Walmart proved you can have: - Affordable prices - Premium ingredients - Unique flavors - Beautiful packaging All at once. The Execution: - 300+ products in Year 1 - Bronze-cut pasta from Italy for $1.97 - Creamy Corn Jalapeño Chowder for under $4 - Plant-based oat milk ice cream that "you won't believe is plant-based" - Hot Honey Seasoning under $3 78% plan to repurchase. 46% say it's better value than other brands. 37% say it's healthier. 34% say higher-quality ingredients. Walmart didn't just launch a product... They repositioned an entire category. What I Learned From This: Stop assuming "premium" means "expensive." Your competitors think they have to choose: - Cheap OR high-quality - Mass market OR exclusive - Accessible OR aspirational Bettergoods proved you can be ALL of it. The framework: 1. Find the assumption everyone makes about your category 2. Do the opposite 3. Package it so both markets see what they want 4. Make the product legitimately good I've seen many companies go cheap to win on price. Or premium to win on quality. Walmart said fuck that and won on BOTH. By making a $3 product feel like $15. And now this isn't just working for Walmart. Target launched Dealworthy (same strategy, different execution). Second-fastest growing private label over the past year. The era of "cheap = low quality" is dead. Smart brands are premiumizing their low-cost offers. Dumb brands are still trying to be the "affordable alternative." What are you?
1:14

Walmart's genius private label strategy literally KILLED name brands... By doing the exact OPPOSITE of what everyone said you should do. In April 2024, they launched Bettergoods - a "cheap" store brand. But instead of targeting broke customers, they went after RICH ones. 28% of U.S. households bought it in one year. $500 million in sales. 40% came back for more. It's now the fastest-growing private label brand in America. This is the most insane private label play in retail history... The Old Playbook (that everyone follows): - Private labels = cheaper version of name brands - Target price-conscious shoppers - Make it look "good enough" - Compete on cost alone Great Value, Kirkland, Equate - they all did this. And it worked. For the bottom 80% of customers. But Walmart saw something nobody else did. The Problem With Traditional Private Labels: High-income shoppers won't touch them. They associate store brands with "lower quality." Even when the product is IDENTICAL, the perception kills sales. So Walmart asked a different question: "What if we made a private label that high-income people actually WANT?" The Bettergoods Strategy: - Most items under $5 (same as Great Value) - But packaging looks like a $15 boutique product - "Chef-inspired" flavors - Trend-forward ingredients (oat milk ice cream, hot honey seasoning, plant-based everything) - Totally unique products - NOT knock-offs of name brands And here's the genius part: They're selling to TWO completely different customers with the SAME product. Low-income shoppers: "I'm treating myself without breaking the bank" High-income shoppers: "This is premium quality at an amazing price" The numbers prove it worked: High-income households are 20% MORE likely to buy Bettergoods than Great Value. Low-income households are 34% more likely to buy Bettergoods than other premium brands. They captured BOTH ends of the market. With the same $3.44 pint of ice cream. Why This Killed Name Brands: Before Bettergoods, high-income Walmart shoppers bought Ben & Jerry's, Häagen-Dazs, Talenti. Now they buy Bettergoods. Same quality. Half the price. Trendier flavors. The name brands lost their premium positioning overnight. Because Walmart proved you can have: - Affordable prices - Premium ingredients - Unique flavors - Beautiful packaging All at once. The Execution: - 300+ products in Year 1 - Bronze-cut pasta from Italy for $1.97 - Creamy Corn Jalapeño Chowder for under $4 - Plant-based oat milk ice cream that "you won't believe is plant-based" - Hot Honey Seasoning under $3 78% plan to repurchase. 46% say it's better value than other brands. 37% say it's healthier. 34% say higher-quality ingredients. Walmart didn't just launch a product... They repositioned an entire category. What I Learned From This: Stop assuming "premium" means "expensive." Your competitors think they have to choose: - Cheap OR high-quality - Mass market OR exclusive - Accessible OR aspirational Bettergoods proved you can be ALL of it. The framework: 1. Find the assumption everyone makes about your category 2. Do the opposite 3. Package it so both markets see what they want 4. Make the product legitimately good I've seen many companies go cheap to win on price. Or premium to win on quality. Walmart said fuck that and won on BOTH. By making a $3 product feel like $15. And now this isn't just working for Walmart. Target launched Dealworthy (same strategy, different execution). Second-fastest growing private label over the past year. The era of "cheap = low quality" is dead. Smart brands are premiumizing their low-cost offers. Dumb brands are still trying to be the "affordable alternative." What are you?

Ricardo

932,767 Aufrufe • vor 9 Monaten

The Cost of Intelligence is Heading to Zero | Hyperspace P2P Distributed Cache We present to you our breakthrough cross-domain work across AI, distributed systems, cryptography, game theory to solve the primary structural inefficiency at the heart of AI infrastructure: most inference is redundant. Google has reported that only 15% of daily searches are truly novel. The rest are repeats or close variants. LLM inference inherits this same power-law distribution. Enterprise chatbots see 70-80% of queries fall into a handful of intent categories. System prompts are identical across 100% of requests within an application. The KV attention state for "You are a helpful assistant" has been computed billions of times, on millions of GPUs, identically. And yet every AI lab, every startup, every self-hosted deployment - computes and caches these results independently. There is no shared layer. No global memory. Every provider pays the full compute cost for every query, even when the answer already exists somewhere in the network. This is the problem Hyperspace solves where distributed cache operates at three levels, each catching a different class of redundancy: 1. Response cache Same prompt, same model, same parameters - instant cached response from any node in the network. SHA-256 hash lookup via DHT, with cryptographic cache proofs linking every response to its original inference execution. No trust required. Fetchers re-announce as providers, so popular responses replicate naturally across more nodes. 2. KV prefix cache Same system prompt tokens - skip the most expensive part of inference entirely. Prefill (computing Key-Value attention states) is deterministic: same model plus same tokens always produces identical KV state. The network caches these states using erasure coding and distributes them via the routing network. New questions that share a common prefix resume generation from cached state instead of recomputing from scratch. 3. Routing to cached nodes Instead of transferring KV state across the network for every request, Hyperspace routes the request to the node that already has the state loaded in VRAM. The request goes to the cache, not the cache to the request. Together, these three layers mean that 70-90% of inference requests at network scale never require full GPU computation. This work doesn't exist in isolation. It builds on research from across the industry: SGLang's RadixAttention demonstrated that automatic prefix sharing can yield up to 5x speedup on structured LLM workloads. Moonshot AI's Mooncake built an entire KV-cache-centric disaggregated architecture for production serving at Kimi. Anthropic, OpenAI, and Google all launched prompt caching products in 2024 - priced at 50-90% discounts - because system prompt reuse is so pervasive that it changes the economics of inference. What all of these systems share is a common limitation: they operate within a single organization's infrastructure. SGLang caches prefixes within one server. Mooncake disaggregates KV cache within one datacenter. Anthropic's prompt caching works within one API provider's fleet. None of them can share cached state across organizational boundaries. Hyperspace removes this boundary. The cache is global. A response computed by a node in Tokyo is immediately available to a node in Berlin. A KV prefix state generated for Qwen-32B on one machine is verifiable and reusable by any other machine running the same model. The routing network provides the delivery guarantees, the erasure coding provides the redundancy, and the cache proofs provide the trust. What this means for the cost of intelligence Big AI labs scale linearly: twice the users means twice the GPU spend. Every query is a cost center. Their internal caching helps, but it's siloed - Lab A's cache can't serve Lab B's users, and neither can serve a self-hosted Llama deployment. Hyperspace scales sub-linearly. Every new node that joins the network adds to the global cache. Every inference result enriches the cache for all future requests. The cache hit rate rises with network size because query distributions follow a power law - the most common questions are asked exponentially more often than rare ones. The implication is simple: as the network grows, the effective cost per inference drops. Not linearly. Logarithmically. At 10 million nodes, we estimate 75-90% of all inference requests can be served from cache, eliminating 400,000+ MWh of energy consumption per year and avoiding over 200,000 tons of CO2 emissions. The first person to ask a question pays the compute cost. Everyone after them gets the answer for free, with cryptographic proof that it's authentic. Training is competitive. Inference is shared Open-weight models are converging on quality with closed models. Labs will continue to differentiate on training - data curation, architecture innovation, RLHF tuning. That's where the real intellectual property lives. But inference is a commodity. Two copies of Qwen-32B running the same prompt produce the same KV state and the same response, byte for byte, regardless of whose GPU runs the matrix multiplication. There is no moat in multiplying matrices. The moat is in training the weights. A global distributed cache makes this separation explicit. It doesn't matter who trained the model. Once the weights are open, the inference cost approaches zero at scale - because the network remembers every answer and can prove it's correct. No lab, no matter how well-funded, can match this. They cannot share caches across competitors. They scale linearly. The network scales logarithmically. The marginal cost of intelligence approaches zero. That's the endgame.
0:23

The Cost of Intelligence is Heading to Zero | Hyperspace P2P Distributed Cache We present to you our breakthrough cross-domain work across AI, distributed systems, cryptography, game theory to solve the primary structural inefficiency at the heart of AI infrastructure: most inference is redundant. Google has reported that only 15% of daily searches are truly novel. The rest are repeats or close variants. LLM inference inherits this same power-law distribution. Enterprise chatbots see 70-80% of queries fall into a handful of intent categories. System prompts are identical across 100% of requests within an application. The KV attention state for "You are a helpful assistant" has been computed billions of times, on millions of GPUs, identically. And yet every AI lab, every startup, every self-hosted deployment - computes and caches these results independently. There is no shared layer. No global memory. Every provider pays the full compute cost for every query, even when the answer already exists somewhere in the network. This is the problem Hyperspace solves where distributed cache operates at three levels, each catching a different class of redundancy: 1. Response cache Same prompt, same model, same parameters - instant cached response from any node in the network. SHA-256 hash lookup via DHT, with cryptographic cache proofs linking every response to its original inference execution. No trust required. Fetchers re-announce as providers, so popular responses replicate naturally across more nodes. 2. KV prefix cache Same system prompt tokens - skip the most expensive part of inference entirely. Prefill (computing Key-Value attention states) is deterministic: same model plus same tokens always produces identical KV state. The network caches these states using erasure coding and distributes them via the routing network. New questions that share a common prefix resume generation from cached state instead of recomputing from scratch. 3. Routing to cached nodes Instead of transferring KV state across the network for every request, Hyperspace routes the request to the node that already has the state loaded in VRAM. The request goes to the cache, not the cache to the request. Together, these three layers mean that 70-90% of inference requests at network scale never require full GPU computation. This work doesn't exist in isolation. It builds on research from across the industry: SGLang's RadixAttention demonstrated that automatic prefix sharing can yield up to 5x speedup on structured LLM workloads. Moonshot AI's Mooncake built an entire KV-cache-centric disaggregated architecture for production serving at Kimi. Anthropic, OpenAI, and Google all launched prompt caching products in 2024 - priced at 50-90% discounts - because system prompt reuse is so pervasive that it changes the economics of inference. What all of these systems share is a common limitation: they operate within a single organization's infrastructure. SGLang caches prefixes within one server. Mooncake disaggregates KV cache within one datacenter. Anthropic's prompt caching works within one API provider's fleet. None of them can share cached state across organizational boundaries. Hyperspace removes this boundary. The cache is global. A response computed by a node in Tokyo is immediately available to a node in Berlin. A KV prefix state generated for Qwen-32B on one machine is verifiable and reusable by any other machine running the same model. The routing network provides the delivery guarantees, the erasure coding provides the redundancy, and the cache proofs provide the trust. What this means for the cost of intelligence Big AI labs scale linearly: twice the users means twice the GPU spend. Every query is a cost center. Their internal caching helps, but it's siloed - Lab A's cache can't serve Lab B's users, and neither can serve a self-hosted Llama deployment. Hyperspace scales sub-linearly. Every new node that joins the network adds to the global cache. Every inference result enriches the cache for all future requests. The cache hit rate rises with network size because query distributions follow a power law - the most common questions are asked exponentially more often than rare ones. The implication is simple: as the network grows, the effective cost per inference drops. Not linearly. Logarithmically. At 10 million nodes, we estimate 75-90% of all inference requests can be served from cache, eliminating 400,000+ MWh of energy consumption per year and avoiding over 200,000 tons of CO2 emissions. The first person to ask a question pays the compute cost. Everyone after them gets the answer for free, with cryptographic proof that it's authentic. Training is competitive. Inference is shared Open-weight models are converging on quality with closed models. Labs will continue to differentiate on training - data curation, architecture innovation, RLHF tuning. That's where the real intellectual property lives. But inference is a commodity. Two copies of Qwen-32B running the same prompt produce the same KV state and the same response, byte for byte, regardless of whose GPU runs the matrix multiplication. There is no moat in multiplying matrices. The moat is in training the weights. A global distributed cache makes this separation explicit. It doesn't matter who trained the model. Once the weights are open, the inference cost approaches zero at scale - because the network remembers every answer and can prove it's correct. No lab, no matter how well-funded, can match this. They cannot share caches across competitors. They scale linearly. The network scales logarithmically. The marginal cost of intelligence approaches zero. That's the endgame.

Varun

37,555 Aufrufe • vor 5 Monaten