Video wird geladen...

Video konnte nicht geladen werden

Beim Laden dieses Videos ist ein Problem aufgetreten. Dies könnte an einem vorübergehenden Netzwerkproblem liegen oder das Video ist möglicherweise nicht verfügbar.

so we already have an open source alternative to jev... and it's 6-7x faster?! it's a typed-decision classification system: no chat, no generated text, just fast yes/no, scoring, or choice answers. > runs in under 1gb of memory > free on hugging face > runs on a laptop, or... show more

Simplifying AI

38,390 subscribers

45,608 Aufrufe • vor 5 Tagen •via X (Twitter)

Nachrichten & Politik Bildung Wissenschaft & Technologie

Anya Rossi• Live Now

Private livecam show

0 Kommentare

Keine Kommentare verfügbar

Kommentare vom Original-Post werden hier angezeigt

Ähnliche Videos

Trained a Jev-style classifier on Hugging Face Jobs for ~$1.50. It's a 194M GLiNER2 model that suggests task tags for any Hub dataset from its column names and first row, and returns a label with a probability. Zero-shot, GLiNER2's first suggestion matched an owner's tag 10% of the time. After 17 minutes of fine-tuning: 69%. The fine-tuned model runs on a free CPU in about a second. Owners' tags are noisy, so some "wrong" answers are tags the owner left out. The recipe is open: one hf jobs command trains the same kind of model on your own labels. The README example (book titles) runs in ~2 minutes for about $0.02. Demo: Recipe:

Trained a Jev-style classifier on Hugging Face Jobs for ~$1.50. It's a 194M GLiNER2 model that suggests task tags for any Hub dataset from its column names and first row, and returns a label with a probability. Zero-shot, GLiNER2's first suggestion matched an owner's tag 10% of the time. After 17 minutes of fine-tuning: 69%. The fine-tuned model runs on a free CPU in about a second. Owners' tags are noisy, so some "wrong" answers are tags the owner left out. The recipe is open: one hf jobs command trains the same kind of model on your own labels. The README example (book titles) runs in ~2 minutes for about $0.02. Demo: Recipe:

Daniel van Strien

15,499 Aufrufe • vor 3 Tagen

Governments pay millions for this kind of system. Someone just open-sourced it for free. It's called Crucix. It watches the entire world and texts you when something changes. Pulls from 26 live sources every 15 minutes: - Satellite fire detection, flight and vessel tracking - Conflict zones and economic indicators - Live crypto and market prices - Social sentiment from Telegram intel channels It's two-way. Pushes alerts to Telegram and Discord. Text it back, type /brief for a summary. No cloud, no subscription. Runs on your own machine. 100% Free. Open Source.

Governments pay millions for this kind of system. Someone just open-sourced it for free. It's called Crucix. It watches the entire world and texts you when something changes. Pulls from 26 live sources every 15 minutes: - Satellite fire detection, flight and vessel tracking - Conflict zones and economic indicators - Live crypto and market prices - Social sentiment from Telegram intel channels It's two-way. Pushes alerts to Telegram and Discord. Text it back, type /brief for a summary. No cloud, no subscription. Runs on your own machine. 100% Free. Open Source.

Simplifying AI

71,505 Aufrufe • vor 2 Monaten

Jev + GrokBot is the best GTM system I've ever built It just made my GTM x200 CHEAPER and x400 FASTER than what 95% of teams are running... setup takes literally 9 minutes: prompt → GrokBot → Jev routes every candidate → GrokBot opens only survivors → campaign step 1 → open TypeSafe AI, create API key (keep it off chat paste) step 2 → tell GrokBot: store TYPESAFE_API_KEY in the secure field step 3 → prompt GrokBot: install typesafe-sdk on Agent Computer + smoke system_one (Choice) step 4 → point it at a target: "read end to end, pull positioning, ICP, every proof point. no copy yet" step 5 → add skill gtm-router: every candidate from X / LinkedIn / YouTube gets 6 typed questions → CHEAP SKIP / OPEN FULL / FLAG KOL step 6 → let Jev rank hook archetypes by viral rate, not by how often people post them - this is where the whole playbook comes from step 7 → GrokBot assembles the KOL shortlist and the 7-day plan - you stay in control of anything irreversible the result: → 3,412 candidates, X, LinkedIn and YouTube. → 15.7 seconds. $0.41. → me doing the same reading: 6h 12m. 8 grok bots brief it, Jev decides, GrokBot executes. no fine-tuning, no embeddings, no vector db - 6 typed questions and a schema. Setup your GrokBot JEV GTM today, then read the article below to learn how to build an agent team with JEV.

Jev + GrokBot is the best GTM system I've ever built It just made my GTM x200 CHEAPER and x400 FASTER than what 95% of teams are running... setup takes literally 9 minutes: prompt → GrokBot → Jev routes every candidate → GrokBot opens only survivors → campaign step 1 → open TypeSafe AI, create API key (keep it off chat paste) step 2 → tell GrokBot: store TYPESAFE_API_KEY in the secure field step 3 → prompt GrokBot: install typesafe-sdk on Agent Computer + smoke system_one (Choice) step 4 → point it at a target: "read end to end, pull positioning, ICP, every proof point. no copy yet" step 5 → add skill gtm-router: every candidate from X / LinkedIn / YouTube gets 6 typed questions → CHEAP SKIP / OPEN FULL / FLAG KOL step 6 → let Jev rank hook archetypes by viral rate, not by how often people post them - this is where the whole playbook comes from step 7 → GrokBot assembles the KOL shortlist and the 7-day plan - you stay in control of anything irreversible the result: → 3,412 candidates, X, LinkedIn and YouTube. → 15.7 seconds. $0.41. → me doing the same reading: 6h 12m. 8 grok bots brief it, Jev decides, GrokBot executes. no fine-tuning, no embeddings, no vector db - 6 typed questions and a schema. Setup your GrokBot JEV GTM today, then read the article below to learn how to build an agent team with JEV.

Movez

84,874 Aufrufe • vor 6 Tagen

Unlimited-OCR is a 3B parameter model that parses entire 100-page PDFs in one shot — no page chunking, no lost context. +⁠32K context window, reads the whole document in a single pass +⁠93% on standard parsing benchmarks, +6 over baseline +⁠Under 0.11 error rate past page 40 — where every other OCR tool falls apart ⁠+Multilingual out of the box, runs locally via Transformers, Ollama, llama.cpp, Docker +⁠Textract, Google Vision, and Azure Doc Intelligence charge $1.50–$15 per 1,000 pages 1.9M downloads on HuggingFace and most people have never heard of it. This runs on your machine. For free. Forever. Model link:

Unlimited-OCR is a 3B parameter model that parses entire 100-page PDFs in one shot — no page chunking, no lost context. +⁠32K context window, reads the whole document in a single pass +⁠93% on standard parsing benchmarks, +6 over baseline +⁠Under 0.11 error rate past page 40 — where every other OCR tool falls apart ⁠+Multilingual out of the box, runs locally via Transformers, Ollama, llama.cpp, Docker +⁠Textract, Google Vision, and Azure Doc Intelligence charge $1.50–$15 per 1,000 pages 1.9M downloads on HuggingFace and most people have never heard of it. This runs on your machine. For free. Forever. Model link:

0xMarioNawfal

84,500 Aufrufe • vor 2 Monaten

China just dropped an open source model that turns any phone into a 3D scanner you point it at a room and walk through it the whole space builds itself into a 3D point cloud in real time as the camera moves, no LiDAR and no depth sensor anywhere it held a full 13 minute walkthrough at 25,000 frames without the map drifting or collapsing > ~20 fps on a single GPU > works indoors and outdoors from plain footage > the scan normally needs a $50,000 rig and a technician > Apache 2.0 and runs fully offline on your machine repo in the reply

China just dropped an open source model that turns any phone into a 3D scanner you point it at a room and walk through it the whole space builds itself into a 3D point cloud in real time as the camera moves, no LiDAR and no depth sensor anywhere it held a full 13 minute walkthrough at 25,000 frames without the map drifting or collapsing > ~20 fps on a single GPU > works indoors and outdoors from plain footage > the scan normally needs a $50,000 rig and a technician > Apache 2.0 and runs fully offline on your machine repo in the reply

savip.

105,305 Aufrufe • vor 2 Monaten

Depth Anything 3 now runs as pure C++/ggml (ggml) . No Python, no PyTorch, no CUDA toolkit at inference, just one self-contained GGUF. It's faster than PyTorch on CPU! and ties speed on GPU. The CPU win came from the last place..I'd have looked. Quantized GGUF on Hugging Face🤗 Shout out to Georgi Gerganov for ggml (we are building a ggml-world!❤️) and to ByteDance Open Source and Depth Anything 3 authors Bingyi Kang Jun Hao Liew Donny Y. Chen !

Depth Anything 3 now runs as pure C++/ggml (ggml) . No Python, no PyTorch, no CUDA toolkit at inference, just one self-contained GGUF. It's faster than PyTorch on CPU! and ties speed on GPU. The CPU win came from the last place..I'd have looked. Quantized GGUF on Hugging Face🤗 Shout out to Georgi Gerganov for ggml (we are building a ggml-world!❤️) and to ByteDance Open Source and Depth Anything 3 authors Bingyi Kang Jun Hao Liew Donny Y. Chen !

Ettore Di Giacinto

34,915 Aufrufe • vor 3 Monaten

The best open source 3D AI generation model now runs natively in ComfyUI. The workflow runs 100% locally, on just 6GB of VRAM. No more complex setup or dependency errors. TRELLIS.2 and Pixal3D work right out of the box, with one click. Not to mention, it's FULLY COMMERCIAL-SAFE. The team re-implemented and removed every strictly-licensed library (thanks to the legend Kijai). Drop in an image and click run, it's that easy to start. Go play with it. Free workflow below!

The best open source 3D AI generation model now runs natively in ComfyUI. The workflow runs 100% locally, on just 6GB of VRAM. No more complex setup or dependency errors. TRELLIS.2 and Pixal3D work right out of the box, with one click. Not to mention, it's FULLY COMMERCIAL-SAFE. The team re-implemented and removed every strictly-licensed library (thanks to the legend Kijai). Drop in an image and click run, it's that easy to start. Go play with it. Free workflow below!

rob - comfyui

73,572 Aufrufe • vor 22 Tagen

New open source library: kugiri (区切り). Already running in production on several of my sites, now public. It splits text into lines, words and characters, the thing every text reveal needs. The difference is where it cuts. Most splitters guess where a line breaks, and the guess falls apart the moment the text does something real: text-wrap: balance, hyphenation, a drop cap, a float, a link that wraps, RTL, Japanese. kugiri does not guess. It asks the browser where the lines already are and cuts exactly there, so the split text looks pixel for pixel like the text before it. No animation, no scroll watching, no opinions. Every unit gets an index and the library gets out of the way. Animate with CSS, the Web Animations API, GSAP or Motion. One file, no dependencies, about 7 kB gzipped. The demo is also the test suite: fifty situations, each checked against what the browser painted, in Chromium, WebKit and Firefox. Demo: npm: Source:

New open source library: kugiri (区切り). Already running in production on several of my sites, now public. It splits text into lines, words and characters, the thing every text reveal needs. The difference is where it cuts. Most splitters guess where a line breaks, and the guess falls apart the moment the text does something real: text-wrap: balance, hyphenation, a drop cap, a float, a link that wraps, RTL, Japanese. kugiri does not guess. It asks the browser where the lines already are and cuts exactly there, so the split text looks pixel for pixel like the text before it. No animation, no scroll watching, no opinions. Every unit gets an index and the library gets out of the way. Animate with CSS, the Web Animations API, GSAP or Motion. One file, no dependencies, about 7 kB gzipped. The demo is also the test suite: fifty situations, each checked against what the browser painted, in Chromium, WebKit and Firefox. Demo: npm: Source:

Edoardo Lunardi

42,898 Aufrufe • vor 21 Tagen

THIS ENGINEER SKIPPED 7,862 WORDS OF AI ROADMAP AND SHIPPED ON A WEEKEND a Pi Zero 2W, an IMX500 camera and a solar panel, that's the whole stack inference runs on-chip so there's no cloud, no latency and no monthly bill Spotted Dove at 0.95, King Parrot at 0.88, runs forever on sun code is open source and the 3D files are free, you just print the case and buy the parts and it works better than most paid wildlife monitoring services out there full video: code (open source): 3D print files: solar power board: save this so you actually have everything in one place when you decide to build it

THIS ENGINEER SKIPPED 7,862 WORDS OF AI ROADMAP AND SHIPPED ON A WEEKEND a Pi Zero 2W, an IMX500 camera and a solar panel, that's the whole stack inference runs on-chip so there's no cloud, no latency and no monthly bill Spotted Dove at 0.95, King Parrot at 0.88, runs forever on sun code is open source and the 3D files are free, you just print the case and buy the parts and it works better than most paid wildlife monitoring services out there full video: code (open source): 3D print files: solar power board: save this so you actually have everything in one place when you decide to build it

leopardracer

63,096 Aufrufe • vor 4 Monaten

I OPEN-SOURCED MY CRYPTO TRADING BOT ON GITHUB. 100% FREE It's called Cabbage. A spot trading bot you can backtest, paper-trade and run live, one command each. Full step-by-step setup guide. It runs on two models: > Astra (GPT-6) built it. Smart but slow. It thinks, writes the code, explains every decision. > Jev runs it. Jev doesn't talk. One word and a probability, in milliseconds. BUY, HOLD or SELL. Under 0.80 it waits. Here's what's inside: 1. Signal engine RSI 14 + EMA 12/26 confluence. It only reads closed candles, so no fake signals from a candle that's still moving. 2. Hard risk rules Fixed ticket size and a percentage stop-loss on every entry. The bot can't size up because it "feels confident". 3. Backtest → HTML report Event-driven backtest that replays history candle by candle, like it's trading live. You get a full HTML report at the end. 4. Paper trading Real exchange data, local fills, zero money at risk. Run one iteration to test, then leave it running. 5. Live trading Real orders on any exchange CCXT supports. Your keys stay in your own .env file. 6. Doctor command Checks your config and exchange connection before you trade. No orders placed. State is saved in SQLite per market and pair, so a restart doesn't wipe your positions. Built on Investing Algorithm Framework 9. Apache-2.0, fork it and make it yours.

I OPEN-SOURCED MY CRYPTO TRADING BOT ON GITHUB. 100% FREE It's called Cabbage. A spot trading bot you can backtest, paper-trade and run live, one command each. Full step-by-step setup guide. It runs on two models: > Astra (GPT-6) built it. Smart but slow. It thinks, writes the code, explains every decision. > Jev runs it. Jev doesn't talk. One word and a probability, in milliseconds. BUY, HOLD or SELL. Under 0.80 it waits. Here's what's inside: 1. Signal engine RSI 14 + EMA 12/26 confluence. It only reads closed candles, so no fake signals from a candle that's still moving. 2. Hard risk rules Fixed ticket size and a percentage stop-loss on every entry. The bot can't size up because it "feels confident". 3. Backtest → HTML report Event-driven backtest that replays history candle by candle, like it's trading live. You get a full HTML report at the end. 4. Paper trading Real exchange data, local fills, zero money at risk. Run one iteration to test, then leave it running. 5. Live trading Real orders on any exchange CCXT supports. Your keys stay in your own .env file. 6. Doctor command Checks your config and exchange connection before you trade. No orders placed. State is saved in SQLite per market and pair, so a restart doesn't wipe your positions. Built on Investing Algorithm Framework 9. Apache-2.0, fork it and make it yours.

sopersone

57,421 Aufrufe • vor 2 Tagen

I was super intrigued when I noticed that JEV returns not just probs for candidate choices, but also a CONFIDENCE SCORE of it's prediction... I was wondering if TypeSafe AI brought back Bayesian Networks or something to learn uncertainty. Turns out - NO. Acc to their docs, confidence is just a derived metric from the classification scores. It carries ZERO extra information. Their demo page shows: confidence = (3 × largest probability − 1) / 2 Confidence is tightly coupled with the classification scores it is measuring the uncertainty for. There is no magic here. The model can produce high probability on a wrong choice and just be confidently wrong.

I was super intrigued when I noticed that JEV returns not just probs for candidate choices, but also a CONFIDENCE SCORE of it's prediction... I was wondering if TypeSafe AI brought back Bayesian Networks or something to learn uncertainty. Turns out - NO. Acc to their docs, confidence is just a derived metric from the classification scores. It carries ZERO extra information. Their demo page shows: confidence = (3 × largest probability − 1) / 2 Confidence is tightly coupled with the classification scores it is measuring the uncertainty for. There is no magic here. The model can produce high probability on a wrong choice and just be confidently wrong.

AVB

66,354 Aufrufe • vor 6 Tagen

China open-sourced a peanut-sized OCR that parses entire 100-page PDFs in one shot.. It's called Unlimited-OCR. Only 3B params. Runs locally. Every other OCR tool chops your doc into pages and loses the thread. this one reads the whole thing in a single pass. → One-shot "long-horizon" parsing (32K context window) → Multilingual, out of the box → 93% on the standard parsing benchmark (+6 over baseline) → <0.11 error rate past 40 pages → Runs 100% locally on your own hardware → Works with Transformers, vLLM, SGLang, Docker, Ollama, llama.cpp Traditional cloud OCR (Textract, Google Vision, Azure Doc Intelligence) costs $1.50–$15 per 1,000 pages. This runs on your machine. For free. Forever. Baidu built it explicitly to push DeepSeek-OCR one step further. Already at 1.9M downloads on Hugging Face and most people have no idea it exists yet. 100% open source.

China open-sourced a peanut-sized OCR that parses entire 100-page PDFs in one shot.. It's called Unlimited-OCR. Only 3B params. Runs locally. Every other OCR tool chops your doc into pages and loses the thread. this one reads the whole thing in a single pass. → One-shot "long-horizon" parsing (32K context window) → Multilingual, out of the box → 93% on the standard parsing benchmark (+6 over baseline) → <0.11 error rate past 40 pages → Runs 100% locally on your own hardware → Works with Transformers, vLLM, SGLang, Docker, Ollama, llama.cpp Traditional cloud OCR (Textract, Google Vision, Azure Doc Intelligence) costs $1.50–$15 per 1,000 pages. This runs on your machine. For free. Forever. Baidu built it explicitly to push DeepSeek-OCR one step further. Already at 1.9M downloads on Hugging Face and most people have no idea it exists yet. 100% open source.

Superman

1,107,031 Aufrufe • vor 2 Monaten

Jev + Kimi K3 for fraud detection! TLDR: Jev classified 100 emails in 1.42 seconds, then I routed the uncertain cases to Kimi K3. The full pipeline got 96/100 correct for only ~$0.07. Video is not sped up, check out the live run! Here was my process: I gave Jev 100 emails to classify (a mix of 50 legit & 50 fraudelent emails). It classified all of them in 1.42 seconds. An underrated feature about Jev is it will give you the confidence score for a classification, so I routed any prediction under 95% confidence to Kimi K3 to be fully sure. 31 emails fell below that threshold. After routing those to Kimi K3, the combined pipeline reached 96% accuracy. The full run took 16 seconds & ~$0.07 in inference costs: - $0.068 from Kimi K3 on Together AI - $0.003 (1/3 of a cent) from Jev on TypeSafe AI. I think this is a really interesting pattern: use a fast specialized model like Jev for the narrow task, then route the uncertain cases to a larger LLM. I feel like this kind of approach could be a game changer for use cases like fraud or anything realtime. You can use the speed & low cost of Jev while having a larger LLM as a fallback to ensure high accuracy.

Jev + Kimi K3 for fraud detection! TLDR: Jev classified 100 emails in 1.42 seconds, then I routed the uncertain cases to Kimi K3. The full pipeline got 96/100 correct for only ~$0.07. Video is not sped up, check out the live run! Here was my process: I gave Jev 100 emails to classify (a mix of 50 legit & 50 fraudelent emails). It classified all of them in 1.42 seconds. An underrated feature about Jev is it will give you the confidence score for a classification, so I routed any prediction under 95% confidence to Kimi K3 to be fully sure. 31 emails fell below that threshold. After routing those to Kimi K3, the combined pipeline reached 96% accuracy. The full run took 16 seconds & ~$0.07 in inference costs: - $0.068 from Kimi K3 on Together AI - $0.003 (1/3 of a cent) from Jev on TypeSafe AI. I think this is a really interesting pattern: use a fast specialized model like Jev for the narrow task, then route the uncertain cases to a larger LLM. I feel like this kind of approach could be a game changer for use cases like fraud or anything realtime. You can use the speed & low cost of Jev while having a larger LLM as a fallback to ensure high accuracy.

Hassan

59,800 Aufrufe • vor 9 Tagen

You've probably scrolled past a dozen posts about Jev this week without anyone telling you what it actually is. It's the first model from TypeSafe, a lab started by one of the researchers behind ChatGPT. It's a decision engine: you give it options, it picks one and tells you how sure it is. It cannot write a single word, and that is the interesting part. Every other AI you use writes. That is the whole interface. So when software needs a plain yes or no, we make a model write a paragraph and then dig the answer back out of it. Fine in a chat window where a human reads it. Bad inside software, where code has to act on it. The bet is that the valuable half was never the writing. It was the deciding. That problem showed up in WordPress years ago, and it is the reason WPVibe works the way it does. The AI does the work. Anything permanent stops and waits, because a delete that skips the trash is not something software should decide on its own.

You've probably scrolled past a dozen posts about Jev this week without anyone telling you what it actually is. It's the first model from TypeSafe, a lab started by one of the researchers behind ChatGPT. It's a decision engine: you give it options, it picks one and tells you how sure it is. It cannot write a single word, and that is the interesting part. Every other AI you use writes. That is the whole interface. So when software needs a plain yes or no, we make a model write a paragraph and then dig the answer back out of it. Fine in a chat window where a human reads it. Bad inside software, where code has to act on it. The bet is that the valuable half was never the writing. It was the deciding. That problem showed up in WordPress years ago, and it is the reason WPVibe works the way it does. The AI does the work. Anything permanent stops and waits, because a delete that skips the trash is not something software should decide on its own.

John Turner

21,029 Aufrufe • vor 9 Tagen

this might actually be the new ugc method i've seen 100s of ai ugc ads lately where it's the same creator in every single one. same voice, same face, same brand colors, never drifts once what they're actually doing is running notch agents with memory they load the brand context one time. cloned voice, reference photo, brand color codes, do-not-say words, taste or they just teach it inline as they chat, and every correction becomes a saved rule then before it generates a single frame, the agent reads every relevant memory first. hard rules, reference assets, language constraints, taste preferences, all of it and it's workspace-wide, so every new ad already knows the brand before you type a word each chat isn't a re-introduction anymore. it's training your brand isn't a prompt. it's permanent context run 30-50 of those a month at a real offer and the volume does the rest punchier cut: the new method nobody's clocked yet run notch agents with memory on load the brand once. cloned voice, reference photo, color codes, do-not-say words, taste. or teach it inline and every correction saves as a rule it reads every relevant memory before it generates a single frame, workspace-wide each chat isn't a re-introduction. it's training. your brand isn't a prompt, it's permanent context run it at a real rate and the volume prints.

Sulfur

15,290 Aufrufe • vor 3 Monaten

COPY is live on Robinhood Chain $COPY CA: 0xac79255f6f404eba14f316e8669d76573a2d7b1e Five walls read a setup before you copy it: how strong the source is, how deep the pool is, whether the market actually turns over, whether the move is still inside the band, and whether there is a fresh mark to trade against. EDGE, DEPTH, TURN, MOMO, PRICE. A wallet you follow buying is not a reason. Most setups die on the first wall, and you get the exact wall that refused it, every time. Every decision is printed. The terminal runs from one process, keeps local state, and half of it needs no key at all. It never predicts a price. It tells you what you would actually be copying. Open source, MIT, runs on your machine. No browser wallet, nothing signs.

COPY is live on Robinhood Chain $COPY CA: 0xac79255f6f404eba14f316e8669d76573a2d7b1e Five walls read a setup before you copy it: how strong the source is, how deep the pool is, whether the market actually turns over, whether the move is still inside the band, and whether there is a fresh mark to trade against. EDGE, DEPTH, TURN, MOMO, PRICE. A wallet you follow buying is not a reason. Most setups die on the first wall, and you get the exact wall that refused it, every time. Every decision is printed. The terminal runs from one process, keeps local state, and half of it needs no key at all. It never predicts a price. It tells you what you would actually be copying. Open source, MIT, runs on your machine. No browser wallet, nothing signs.

Lunar

112,261 Aufrufe • vor 16 Tagen

I explored a further possibility with local models: Qwen3.6 35B A3B + NVIDIA LocateAnything-3B as a local Computer Use agent (proof of concept). In the demo, I asked it to switch my Mac to light mode. It did. Then back to dark. Did that too — finding the right toggle in System Settings, clicking it, and verifying the change itself. It's fully screenshot-based, so no Accessibility API needed. If it's on screen, the agent can see it and act on it. This runs entirely on your own hardware — private, local, built from two small open models.

I explored a further possibility with local models: Qwen3.6 35B A3B + NVIDIA LocateAnything-3B as a local Computer Use agent (proof of concept). In the demo, I asked it to switch my Mac to light mode. It did. Then back to dark. Did that too — finding the right toggle in System Settings, clicking it, and verifying the change itself. It's fully screenshot-based, so no Accessibility API needed. If it's on screen, the agent can see it and act on it. This runs entirely on your own hardware — private, local, built from two small open models.

stevibe

44,248 Aufrufe • vor 3 Monaten

This Chinese developer launched Llama 70B locally on a MacBook on a plane and for a full 11 hours without internet ran client projects. He was sitting by the window on a transatlantic flight with a MacBook Pro M4 with 64 GB of memory. WiFi on board cost $25 for the flight. He declined. No cloud API, no connection to Anthropic or OpenAI servers, no internet at all. Just a local Llama 3.3 70B on bf16 and his own orchestrator script. The model runs through llama.cpp. Generation speed, 71 tokens per second. Context around 60,000 tokens. Memory usage, 48.6 GiB out of 64. Battery at takeoff, 3 hours 21 minutes. And he gave the orchestrator this system prompt before takeoff: "You are an offline orchestrator running on a single MacBook. There is no network. The only resources you have are local files in /Users/dev/work, the Llama 70B inference server at localhost:8080, and a battery budget of 3 hours 21 minutes. Process the queue at /Users/dev/work/queue.jsonl (one client task per line). For each task: draft → run local evals → save artefact to /Users/dev/work/done/. Save context checkpoints every 12 tasks so you can resume after a battery swap. Stop only on empty queue or when battery drops below 5%." So the system knows exactly what resources it is running on. It knows it has no connection to the outside world for the next 11 hours. It knows it has finite memory and a finite battery. It knows the human will not intervene until the plane lands. The system runs in 1 loop. Takes a task from the queue, runs it through inference, saves the artifact, writes a checkpoint. Task after task, just like that. And only when the battery drops below 5% does the orchestrator automatically pause, waits for the laptop to switch to the backup power bank, and continues from the last checkpoint. Here is what the system actually writes in his log during the flight: "saved context checkpoint 8 of 12 (pos_min = 488, pos_max = 50118, size = 62.813 MiB)" "restored context checkpoint (pos_min = 488, pos_max = 50118)" "prompt processing progress: n_tokens = 50 / 60 818" "task 37016 done | tps = 71 s tokens text → /Users/dev/work/done/proposal_westside.md" Outside the window, clouds, blue sky, and no WiFi. On the tray, 1 MacBook, an open terminal on 2 screens, and an inference server on localhost. From what I have observed, this is the cleanest offline AI workflow I have seen in the past year: 11 hours of flight, $0 for WiFi, and the entire client queue closed before landing.

This Chinese developer launched Llama 70B locally on a MacBook on a plane and for a full 11 hours without internet ran client projects. He was sitting by the window on a transatlantic flight with a MacBook Pro M4 with 64 GB of memory. WiFi on board cost $25 for the flight. He declined. No cloud API, no connection to Anthropic or OpenAI servers, no internet at all. Just a local Llama 3.3 70B on bf16 and his own orchestrator script. The model runs through llama.cpp. Generation speed, 71 tokens per second. Context around 60,000 tokens. Memory usage, 48.6 GiB out of 64. Battery at takeoff, 3 hours 21 minutes. And he gave the orchestrator this system prompt before takeoff: "You are an offline orchestrator running on a single MacBook. There is no network. The only resources you have are local files in /Users/dev/work, the Llama 70B inference server at localhost:8080, and a battery budget of 3 hours 21 minutes. Process the queue at /Users/dev/work/queue.jsonl (one client task per line). For each task: draft → run local evals → save artefact to /Users/dev/work/done/. Save context checkpoints every 12 tasks so you can resume after a battery swap. Stop only on empty queue or when battery drops below 5%." So the system knows exactly what resources it is running on. It knows it has no connection to the outside world for the next 11 hours. It knows it has finite memory and a finite battery. It knows the human will not intervene until the plane lands. The system runs in 1 loop. Takes a task from the queue, runs it through inference, saves the artifact, writes a checkpoint. Task after task, just like that. And only when the battery drops below 5% does the orchestrator automatically pause, waits for the laptop to switch to the backup power bank, and continues from the last checkpoint. Here is what the system actually writes in his log during the flight: "saved context checkpoint 8 of 12 (pos_min = 488, pos_max = 50118, size = 62.813 MiB)" "restored context checkpoint (pos_min = 488, pos_max = 50118)" "prompt processing progress: n_tokens = 50 / 60 818" "task 37016 done | tps = 71 s tokens text → /Users/dev/work/done/proposal_westside.md" Outside the window, clouds, blue sky, and no WiFi. On the tray, 1 MacBook, an open terminal on 2 screens, and an inference server on localhost. From what I have observed, this is the cleanest offline AI workflow I have seen in the past year: 11 hours of flight, $0 for WiFi, and the entire client queue closed before landing.

Blaze

1,843,280 Aufrufe • vor 4 Monaten

I built HypeMeter in 4 hours with Jev + Minds. Its best trick is saying no, and deciding what is likely a rug vs real hype. I am giving away an Argonaut NFT to reward Beta testers. Yes, that's you. Every "alpha bot" screams BUY. None of them tell you which cheap listings are cheap for a reason. So I wired two things together: Jev by TypeSafe AI . It does not write essays. It answers typed questions: pick one, score this, yes or no. About a third of a second per decision, cheap enough to judge every cheap listing instead of a shortlist. Minds by Minds by Animoca Brands . Your own AI agent. Tell it your strategy in plain words ("Argonauts under 0.3, grade A or better") and it messages you one digest a day, pings you whenever steals are available. First full sweep: 898 listings across 20 collections, including Robinhood (of course). Calls that survived: one. And that one was my own bug: an "83% edge" that was a 2-item bid read as one. The sanity check now kills those before anyone sees them. That is the product. Most cheap NFTs are traps, and it says so. It also hunts rares priced under what their trait actually sells for. Yesterday it flagged an Argonaut with a 1-in-70 palette, listed at 0.79 ETH two days before the same palette sold for 0.9 and 1.0. No hindsight. Every call is written down the moment it is made, then graded at 24 hours and 7 days. Public scoreboard, losses included. Free while in beta. Sign in with Minds: And yes, the giveaway is real: Argonaut #2764 goes to someone who actually uses it. Every active day is an entry, there is a leaderboard, and signing in before 24 Sept gets you 3 bonus entries. Rules on the site. RT and comment "Jev" for extra entry. Have fun sniping.

I built HypeMeter in 4 hours with Jev + Minds. Its best trick is saying no, and deciding what is likely a rug vs real hype. I am giving away an Argonaut NFT to reward Beta testers. Yes, that's you. Every "alpha bot" screams BUY. None of them tell you which cheap listings are cheap for a reason. So I wired two things together: Jev by TypeSafe AI . It does not write essays. It answers typed questions: pick one, score this, yes or no. About a third of a second per decision, cheap enough to judge every cheap listing instead of a shortlist. Minds by Minds by Animoca Brands . Your own AI agent. Tell it your strategy in plain words ("Argonauts under 0.3, grade A or better") and it messages you one digest a day, pings you whenever steals are available. First full sweep: 898 listings across 20 collections, including Robinhood (of course). Calls that survived: one. And that one was my own bug: an "83% edge" that was a 2-item bid read as one. The sanity check now kills those before anyone sees them. That is the product. Most cheap NFTs are traps, and it says so. It also hunts rares priced under what their trait actually sells for. Yesterday it flagged an Argonaut with a 1-in-70 palette, listed at 0.79 ETH two days before the same palette sold for 0.9 and 1.0. No hindsight. Every call is written down the moment it is made, then graded at 24 hours and 7 days. Public scoreboard, losses included. Free while in beta. Sign in with Minds: And yes, the giveaway is real: Argonaut #2764 goes to someone who actually uses it. Every active day is an entry, there is a leaderboard, and signing in before 24 Sept gets you 3 bonus entries. Rules on the site. RT and comment "Jev" for extra entry. Have fun sniping.

Jesus is Lord | Chev

38,754 Aufrufe • vor 4 Tagen

Last week we introduced Kenji to the button that initiates a video call to an emergency contact. Because he already knows to push buttons, he only needed to learn that it was a paw push not a nose push, and the cue "FaceTime". Today he had his second training session and we moved the button into the living room where it will be used, and he immediately went and pushed it when cued. Then we started to add distance. We did a few runs from within the living room, then a few from the kitchen, and finally a few from the back porch. He did it every single time. Young dogs learn very quickly. We kept the session short and just did a few repetitions at the furthest distance. It's tempting to go further away or into a different room, but it's better to leave that for the next session. Overtraining can take the fun out of it for the dog. Yes, they need to get lots of repetitions in for a behavior to become precise and reliable, but we want to keep it enjoyable and not increase difficulty too quickly. If he's already done several repetitions from a distance and then we ask him to do it from a room further away, then we run the risk of him losing interest in the activity and becoming more interested in some of the distractions he passes along the way. Like his bone and a toy box. When training a new skill, the dog needs lots of wins. It's far better for us to end the session where we did, rather than move to the bedroom and have him decide to stop and check out his bone on his way to the button. In a day or so we will do another session with him doing it once from the living room, a couple of fast repetitions from the back porch with exciting rewards, and then increase the difficulty while he's highly motivated. The session after that we will probably hook it up and introduce the sounds that it makes. He's doing awesome at this. It's not surprising. We know that there's more in his giant noggin than just marshmallows. Even if he's still hoping that button let's him video chat with Pizza Hut. 🎥 Kenji being awesome #Rottweiler #PuppyTraining #KenjiChaos

Last week we introduced Kenji to the button that initiates a video call to an emergency contact. Because he already knows to push buttons, he only needed to learn that it was a paw push not a nose push, and the cue "FaceTime". Today he had his second training session and we moved the button into the living room where it will be used, and he immediately went and pushed it when cued. Then we started to add distance. We did a few runs from within the living room, then a few from the kitchen, and finally a few from the back porch. He did it every single time. Young dogs learn very quickly. We kept the session short and just did a few repetitions at the furthest distance. It's tempting to go further away or into a different room, but it's better to leave that for the next session. Overtraining can take the fun out of it for the dog. Yes, they need to get lots of repetitions in for a behavior to become precise and reliable, but we want to keep it enjoyable and not increase difficulty too quickly. If he's already done several repetitions from a distance and then we ask him to do it from a room further away, then we run the risk of him losing interest in the activity and becoming more interested in some of the distractions he passes along the way. Like his bone and a toy box. When training a new skill, the dog needs lots of wins. It's far better for us to end the session where we did, rather than move to the bedroom and have him decide to stop and check out his bone on his way to the button. In a day or so we will do another session with him doing it once from the living room, a couple of fast repetitions from the back porch with exciting rewards, and then increase the difficulty while he's highly motivated. The session after that we will probably hook it up and introduce the sounds that it makes. He's doing awesome at this. It's not surprising. We know that there's more in his giant noggin than just marshmallows. Even if he's still hoping that button let's him video chat with Pizza Hut. 🎥 Kenji being awesome #Rottweiler #PuppyTraining #KenjiChaos

Team Servicerottie🇨🇦🐕‍🦺🦽

10,152 Aufrufe • vor 1 Monat