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.

I thought my AI agent needed a bigger context window wrong the leak was my shell every time it ran git diff HEAD~5 the model ate 92,000 tokens after ztk 18,000 ls -la src 2,000 tokens became 53 cargo test passing 397 tokens became 21 the fix is a... 260KB Zig binary that sits between your agent and the shell it compresses command output before it hits the model not by summarizing everything by knowing what each command actually needs to preserve > diffs keep changed lines > tests keep failures > ls keeps structure > logs dedupe repeats > JSON and errors pass through untouched 5.8M tokens saved in one 256 command session stop buying bigger context windows while feeding your agent raw terminal sewageshow more

Alpha Batcher

51,749 subscribers

17,516 Aufrufe • vor 3 Monaten •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

0 Kommentare

Keine Kommentare verfügbar

Kommentare vom Original-Post werden hier angezeigt

Ähnliche Videos

Told my OpenClaw agent: "Cut my token spend in half without touching output quality"... Woke up 5 hours later. Done. Here's what changed: Two-tier memory - stops loading the entire context on every single request. Bootstrap holds only the critical rules. Everything else lives in MEMORY md, semantic search pulls it when it's actually needed. The numbers: Before: 8,200 tokens/request = $73/day After: 2,700 tokens/request = $24/day 67% gone, every day, forever. I'm use for copytrade traders: Ran Monte Carlo across 200 iterations - holds up clean. The principle is dead simple: don't load what you don't need right now. Old approach is like reading Polymarket entire market history to price one contract. Semantic search is pulling only the relevant resolution criteria and recent volume. Two weeks post-switch: $1,134 saved Polymarket research agent now processes 3x more markets in the same window Caught 31 mispriced YES/NO spreads before the crowd adjusted - faster responses meant faster entries One architecture change, one config tweak. You are either paying for tokens you don't use. Or you are not.

Told my OpenClaw agent: "Cut my token spend in half without touching output quality"... Woke up 5 hours later. Done. Here's what changed: Two-tier memory - stops loading the entire context on every single request. Bootstrap holds only the critical rules. Everything else lives in MEMORY md, semantic search pulls it when it's actually needed. The numbers: Before: 8,200 tokens/request = $73/day After: 2,700 tokens/request = $24/day 67% gone, every day, forever. I'm use for copytrade traders: Ran Monte Carlo across 200 iterations - holds up clean. The principle is dead simple: don't load what you don't need right now. Old approach is like reading Polymarket entire market history to price one contract. Semantic search is pulling only the relevant resolution criteria and recent volume. Two weeks post-switch: $1,134 saved Polymarket research agent now processes 3x more markets in the same window Caught 31 mispriced YES/NO spreads before the crowd adjusted - faster responses meant faster entries One architecture change, one config tweak. You are either paying for tokens you don't use. Or you are not.

Lunar

37,539 Aufrufe • vor 5 Monaten

If your AI replies instantly… it’s probably not building. Replit’s Agent just became your personal engineer. And it’s changing how we think about building apps I saw it complete a full feature build from one prompt. Try it here: • Understood the context of what I wanted • Wrote the code without micromanagement • Tested and refined on its own • Shipped a working dashboard overnight All while I stepped away. This isn’t just “AI assistance.” It’s delegation to an Agent that actually thinks. Here’s what actually happened: 1. Typed one clear prompt 2. Agent ran longer without babysitting 3. Built features autonomously 4. Debugged in the background 5. Delivered a working result by morning It wasn’t just fast output. It built something real and functional. This is the future we’ve been waiting for: ☑︎ AI that doesn’t stop at a single reply ☑︎ Agents that understand context, not just commands ☑︎ Work that gets done while you focus on vision We’re moving from “AI that chats” to “AI that builds.” Replit Agent just showed us what’s possible. Try it now: Learning something new? Repost ♻️ so others can too.

If your AI replies instantly… it’s probably not building. Replit’s Agent just became your personal engineer. And it’s changing how we think about building apps I saw it complete a full feature build from one prompt. Try it here: • Understood the context of what I wanted • Wrote the code without micromanagement • Tested and refined on its own • Shipped a working dashboard overnight All while I stepped away. This isn’t just “AI assistance.” It’s delegation to an Agent that actually thinks. Here’s what actually happened: 1. Typed one clear prompt 2. Agent ran longer without babysitting 3. Built features autonomously 4. Debugged in the background 5. Delivered a working result by morning It wasn’t just fast output. It built something real and functional. This is the future we’ve been waiting for: ☑︎ AI that doesn’t stop at a single reply ☑︎ Agents that understand context, not just commands ☑︎ Work that gets done while you focus on vision We’re moving from “AI that chats” to “AI that builds.” Replit Agent just showed us what’s possible. Try it now: Learning something new? Repost ♻️ so others can too.

Muhammad Ayan

58,651 Aufrufe • vor 11 Monaten

whoever leaked this has bigger balls than sense Google Research and MIT ran the same agent jobs 260 different ways for Nature last month: they held the prompts, the tools and the compute budget identical and moved nothing but the wiring between the agents, and the same work swung from 70% worse than a single agent to 80.8% better, averaging out at 0.0% i ran my own single agent against the task list first and it cleared 6 of 10 alone, already past the line where a crew starts subtracting this is Graph Engineering, the layer that decides whether a crew is worth 80% more or 70% less, and it installs into the agent you already pay for: - score your solo agent on the real task first: above roughly 45% success that study predicts zero to negative returns from any crew you put around it - under that line, put one supervisor over the fan out: crews with no correction step amplified their own errors to 17.2x the single agent rate, supervised aggregation held it to 4.4x - give every worker one output and let none of them read a peer's draft, so a wrong step reaches the supervisor instead of four other agents - run the comparison again after every model upgrade, because a better model raises your baseline and a higher baseline is what makes a crew stop paying - keep the single agent alive as the control, the only number that says the wiring is earning its calls turns out the shape does not travel: the biggest win came off a finance task under one supervisor and the worst collapse off a planning task with independent agents my position, and it is the arguable one: a crew is a bet on your own diagram, and the model you pick moves that bet less than one arrow does bookmark this, the three moves that draw those arrows before you pay for one extra call are in the post below ↓

whoever leaked this has bigger balls than sense Google Research and MIT ran the same agent jobs 260 different ways for Nature last month: they held the prompts, the tools and the compute budget identical and moved nothing but the wiring between the agents, and the same work swung from 70% worse than a single agent to 80.8% better, averaging out at 0.0% i ran my own single agent against the task list first and it cleared 6 of 10 alone, already past the line where a crew starts subtracting this is Graph Engineering, the layer that decides whether a crew is worth 80% more or 70% less, and it installs into the agent you already pay for: - score your solo agent on the real task first: above roughly 45% success that study predicts zero to negative returns from any crew you put around it - under that line, put one supervisor over the fan out: crews with no correction step amplified their own errors to 17.2x the single agent rate, supervised aggregation held it to 4.4x - give every worker one output and let none of them read a peer's draft, so a wrong step reaches the supervisor instead of four other agents - run the comparison again after every model upgrade, because a better model raises your baseline and a higher baseline is what makes a crew stop paying - keep the single agent alive as the control, the only number that says the wiring is earning its calls turns out the shape does not travel: the biggest win came off a finance task under one supervisor and the worst collapse off a planning task with independent agents my position, and it is the arguable one: a crew is a bet on your own diagram, and the model you pick moves that bet less than one arrow does bookmark this, the three moves that draw those arrows before you pay for one extra call are in the post below ↓

Argona

858,655 Aufrufe • vor 1 Tag

Most AI agent setups treat every message the same. Simple question? top-tier model. complex task? top-tier model. Your token bill just keeps climbing. I tested OpenSquilla this week on a real document drafting workflow, and the routing caught me off guard. It judges each message's complexity locally, then picks the model tier that fits. Simple tasks go to cheaper models. Complex ones still get the heavy lifting done. You're not paying reasoning tokens for a "hello." I ran a longer workflow, and the context didn't collapse the way it usually does. It distills important information before compression, so you're not starting from scratch mid-session. If you run agents regularly, the bill adds up faster than you think. This is built specifically for that problem. They're running the 10M Token Bill Challenge right now. worth joining if you want to see what smart routing actually saves you in practice. #10MTokenChallenge OpenSquilla

Most AI agent setups treat every message the same. Simple question? top-tier model. complex task? top-tier model. Your token bill just keeps climbing. I tested OpenSquilla this week on a real document drafting workflow, and the routing caught me off guard. It judges each message's complexity locally, then picks the model tier that fits. Simple tasks go to cheaper models. Complex ones still get the heavy lifting done. You're not paying reasoning tokens for a "hello." I ran a longer workflow, and the context didn't collapse the way it usually does. It distills important information before compression, so you're not starting from scratch mid-session. If you run agents regularly, the bill adds up faster than you think. This is built specifically for that problem. They're running the 10M Token Bill Challenge right now. worth joining if you want to see what smart routing actually saves you in practice. #10MTokenChallenge OpenSquilla

Parul Gautam

26,685 Aufrufe • vor 3 Monaten

🚨 Do you understand what Claude just quietly dropped while everyone was distracted? 1 million tokens. Let me explain what that actually means because the number alone doesn't hit right. > A senior engineer joins a company and spends 3 to 6 months just reading code.. Understanding how things connect. Learning where the bugs hide. Why that one file nobody touches exists. It takes months because a codebase is massive and human memory is small. > Claude just loaded the entire thing in one prompt. 30 seconds. Every file, Every function, Every line. All of it. Sitting in memory like it's been working there for years. And it scored highest among every single frontier model. Not GPT.. Not Gemini, Nobody. > Yesterday Amazon's AI nuked production because it couldn't see the full picture - it made a decision with partial context and deleted everything. Today an AI can hold 1 million tokens of context at once. That's the fix. That's the "before and after" moment for AI coding. > 600 images in one request. Entire PDFs. Full repos. And they dropped it on a Friday on all plans like it was a patch note. The scariest AI updates aren't the ones with press conferences. They're the ones that drop in a tweet at 6pm and change everything by Monday morning.

🚨 Do you understand what Claude just quietly dropped while everyone was distracted? 1 million tokens. Let me explain what that actually means because the number alone doesn't hit right. > A senior engineer joins a company and spends 3 to 6 months just reading code.. Understanding how things connect. Learning where the bugs hide. Why that one file nobody touches exists. It takes months because a codebase is massive and human memory is small. > Claude just loaded the entire thing in one prompt. 30 seconds. Every file, Every function, Every line. All of it. Sitting in memory like it's been working there for years. And it scored highest among every single frontier model. Not GPT.. Not Gemini, Nobody. > Yesterday Amazon's AI nuked production because it couldn't see the full picture - it made a decision with partial context and deleted everything. Today an AI can hold 1 million tokens of context at once. That's the fix. That's the "before and after" moment for AI coding. > 600 images in one request. Entire PDFs. Full repos. And they dropped it on a Friday on all plans like it was a patch note. The scariest AI updates aren't the ones with press conferences. They're the ones that drop in a tweet at 6pm and change everything by Monday morning.

Tuki

206,260 Aufrufe • vor 5 Monaten

Don't train the model, evolve the harness. I read a brilliant blog post from Hugging Face where they took a frozen open model scoring 0% on a hard legal agent benchmark, left its weights alone, and let an automated loop rewrite only the code around it. That code layer is the harness, the runtime wrapper that feeds the model context, runs its tool calls, and decides when a run ends. By the time the loop finished, the system had essentially matched Sonnet 4.6 on the benchmark's headline metric, at roughly 7x lower cost per task. Zero weights changed. The gain existed because of where the model was failing. The judge only grades files saved in the right place under the exact requested filename, and the model kept doing the legal analysis correctly, then saving it under the wrong name, dropping it in a scratch folder, or never writing it at all. So the 0% was never measuring legal reasoning. It was measuring the harness. Hand-tuning that layer is slow and model-specific, so they automated it. A Claude proposer adds exactly one mechanism per iteration, and an outer loop keeps it only if it clearly beats the current best, so accepted mechanisms compound. What the loop discovered says a lot about where agents actually fail. → The biggest single gain was file handling, not intelligence. An automatic step that lands the deliverable exactly where the judge expects it beat every prompt change, with zero extra model tokens. → Code fixes transferred across models, prompt playbooks did not. The same harness lifted a smaller model from the same family by 14 points, but the tuned prompts hurt a different model family on tasks it could already finish. → The harness mattered more than anything else. Same model, same judge, same tasks, and five different harnesses scored anywhere between 3.5% and 80.1%. The gains do eventually flatten, and the remaining misses look like real capability gaps. At some point the wrapper runs out of tricks and the model has to carry the work. But the lesson holds. A benchmark score measures the model and its harness together, and until the harness is fixed, it's impossible to know which one failed. I highly recommend reading this: I also wrote a deep dive on agent harness engineering a while back, covering the orchestration loop, tools, memory, context management, and everything that turns a stateless LLM into a capable agent. The article is quoted below.

Don't train the model, evolve the harness. I read a brilliant blog post from Hugging Face where they took a frozen open model scoring 0% on a hard legal agent benchmark, left its weights alone, and let an automated loop rewrite only the code around it. That code layer is the harness, the runtime wrapper that feeds the model context, runs its tool calls, and decides when a run ends. By the time the loop finished, the system had essentially matched Sonnet 4.6 on the benchmark's headline metric, at roughly 7x lower cost per task. Zero weights changed. The gain existed because of where the model was failing. The judge only grades files saved in the right place under the exact requested filename, and the model kept doing the legal analysis correctly, then saving it under the wrong name, dropping it in a scratch folder, or never writing it at all. So the 0% was never measuring legal reasoning. It was measuring the harness. Hand-tuning that layer is slow and model-specific, so they automated it. A Claude proposer adds exactly one mechanism per iteration, and an outer loop keeps it only if it clearly beats the current best, so accepted mechanisms compound. What the loop discovered says a lot about where agents actually fail. → The biggest single gain was file handling, not intelligence. An automatic step that lands the deliverable exactly where the judge expects it beat every prompt change, with zero extra model tokens. → Code fixes transferred across models, prompt playbooks did not. The same harness lifted a smaller model from the same family by 14 points, but the tuned prompts hurt a different model family on tasks it could already finish. → The harness mattered more than anything else. Same model, same judge, same tasks, and five different harnesses scored anywhere between 3.5% and 80.1%. The gains do eventually flatten, and the remaining misses look like real capability gaps. At some point the wrapper runs out of tricks and the model has to carry the work. But the lesson holds. A benchmark score measures the model and its harness together, and until the harness is fixed, it's impossible to know which one failed. I highly recommend reading this: I also wrote a deep dive on agent harness engineering a while back, covering the orchestration loop, tools, memory, context management, and everything that turns a stateless LLM into a capable agent. The article is quoted below.

Akshay 🚀

244,567 Aufrufe • vor 1 Monat

This broke my mental model of game dev 💀 2.5 hours → fully playable ‘Worms’ clone. Built with Hermes agent by Nous Research Here’s what made that speed possible: Hermes used ‘Persistent Shell’ mode, which ensured it didn't forget its current folder or active tools. This allowed it to work smoothly, without the distraction of constantly having to recall where it left off last time. To optimize the workflow, the agent moved beyond linear execution and parallelized the workload. It spawned isolated subagents while executing multiple independent tool calls via ThreadPoolExecutor. Like, one subagent wrote Python RPC scripts for the projectile physics while another utilized vision tools for character sprites. When the complex terrain logic required debugging, the agent used filesystem checkpoints and the /rollback command to instantly return to a stable state. To fix UI bugs, it attached to a live Chrome instance via CDP (/browser connect), fixing rendering issues in real-time. The agent’s built-in learning loop was active from the very beginning. By the time the game was finished, this continuous process allowed the agent to autonomously convert the physics logic into a custom skill. This logic is now a permanent plugin file in the agent's plugin architecture, making the physics engine a native capability that the agent can reuse for future projects. Follow War_v3_FINALE.exe for updates!

This broke my mental model of game dev 💀 2.5 hours → fully playable ‘Worms’ clone. Built with Hermes agent by Nous Research Here’s what made that speed possible: Hermes used ‘Persistent Shell’ mode, which ensured it didn't forget its current folder or active tools. This allowed it to work smoothly, without the distraction of constantly having to recall where it left off last time. To optimize the workflow, the agent moved beyond linear execution and parallelized the workload. It spawned isolated subagents while executing multiple independent tool calls via ThreadPoolExecutor. Like, one subagent wrote Python RPC scripts for the projectile physics while another utilized vision tools for character sprites. When the complex terrain logic required debugging, the agent used filesystem checkpoints and the /rollback command to instantly return to a stable state. To fix UI bugs, it attached to a live Chrome instance via CDP (/browser connect), fixing rendering issues in real-time. The agent’s built-in learning loop was active from the very beginning. By the time the game was finished, this continuous process allowed the agent to autonomously convert the physics logic into a custom skill. This logic is now a permanent plugin file in the agent's plugin architecture, making the physics engine a native capability that the agent can reuse for future projects. Follow War_v3_FINALE.exe for updates!

Javier

37,874 Aufrufe • vor 4 Monaten

Another WTF moment. A developer just open-sourced a coding agent harness that boots 245x faster than Claude Code. It's called jcode. You launch it and the first frame renders in 14 milliseconds. Claude Code takes 3,436. One active session uses 27.8 MB of RAM. Claude Code uses 386.6. Run ten sessions in parallel and jcode holds at 117 MB while OpenCode swells to 3.2 GB. Each agent has a semantic memory graph instead of a scratchpad. Every turn gets embedded as a vector. The graph is queried on every turn for related memories, and a sideagent verifies the hits before injecting them into context. Consolidation runs in the background to check for stale or conflicting facts. No manual /remember calls. No token burn on lookup tools. The provider list is 30+ deep. Claude, ChatGPT, Gemini, GitHub Copilot, Azure, OpenRouter, DeepSeek, Groq, Mistral, Perplexity, Fireworks, Ollama, LM Studio, and any OpenAI-compatible endpoint you point it at. Ran out of tokens on your first ChatGPT Pro sub? /account swaps to the second. Then there's Swarm. Spawn two agents in the same repo and the server manages them. When agent A edits a file agent B has been reading, agent B gets pinged and can check the diff. Agents can DM each other, broadcast to the room, or spawn their own worker teams for parallel tasks. Groups, channels, and completion statuses are handled automatically. The UI has live side panels that render mermaid diagrams inline. To make it fast, the author wrote a Rust mermaid renderer 1800x faster than the JavaScript one, then wrote a custom terminal called Handterm because no existing terminal could do smooth partial-line scrolling. Self-dev mode is where it gets wild. Tell your agent to enter self-dev and it starts editing jcode's own source code, rebuilds the binary, reloads it live, and keeps working across your existing sessions. You can also resume broken sessions from Claude Code, Codex, OpenCode, or pi directly inside jcode. Anthropic's cache goes cold at the 5-minute mark and you're staring down a big cache miss on your next turn? The UI warns you before you spend the tokens. Written in Rust. MIT licensed. Runs on macOS, Windows, Linux, and Termux. Sitting at 11.2k stars with a native iOS app coming.

Another WTF moment. A developer just open-sourced a coding agent harness that boots 245x faster than Claude Code. It's called jcode. You launch it and the first frame renders in 14 milliseconds. Claude Code takes 3,436. One active session uses 27.8 MB of RAM. Claude Code uses 386.6. Run ten sessions in parallel and jcode holds at 117 MB while OpenCode swells to 3.2 GB. Each agent has a semantic memory graph instead of a scratchpad. Every turn gets embedded as a vector. The graph is queried on every turn for related memories, and a sideagent verifies the hits before injecting them into context. Consolidation runs in the background to check for stale or conflicting facts. No manual /remember calls. No token burn on lookup tools. The provider list is 30+ deep. Claude, ChatGPT, Gemini, GitHub Copilot, Azure, OpenRouter, DeepSeek, Groq, Mistral, Perplexity, Fireworks, Ollama, LM Studio, and any OpenAI-compatible endpoint you point it at. Ran out of tokens on your first ChatGPT Pro sub? /account swaps to the second. Then there's Swarm. Spawn two agents in the same repo and the server manages them. When agent A edits a file agent B has been reading, agent B gets pinged and can check the diff. Agents can DM each other, broadcast to the room, or spawn their own worker teams for parallel tasks. Groups, channels, and completion statuses are handled automatically. The UI has live side panels that render mermaid diagrams inline. To make it fast, the author wrote a Rust mermaid renderer 1800x faster than the JavaScript one, then wrote a custom terminal called Handterm because no existing terminal could do smooth partial-line scrolling. Self-dev mode is where it gets wild. Tell your agent to enter self-dev and it starts editing jcode's own source code, rebuilds the binary, reloads it live, and keeps working across your existing sessions. You can also resume broken sessions from Claude Code, Codex, OpenCode, or pi directly inside jcode. Anthropic's cache goes cold at the 5-minute mark and you're staring down a big cache miss on your next turn? The UI warns you before you spend the tokens. Written in Rust. MIT licensed. Runs on macOS, Windows, Linux, and Termux. Sitting at 11.2k stars with a native iOS app coming.

Brady Long

204,606 Aufrufe • vor 19 Tagen

this is worth more than most five figure courses 16 claude agents audit an entire repo at once, a second fleet re-checks every finding on fresh context, and the whole thing runs off one diagram instead of a prompt i ran it against my own code and got back 11 endpoints where i never checked who was logged in, 3 of which the verifier threw out before they ever reached me this is Graph Engineering, the layer above prompting, and it runs on the agent you already pay for: - write your plan out, then ask one question at every "and then": does the next step actually read what the previous one produced - the seams that fail that question were never dependencies, so those jobs run at the same time - the arrows that survive are your real edges, and the longest chain of them is your floor that no number of agents shortens - want it faster, cut a false edge instead of adding a worker - fan the independent work out, one agent per item, no shared state between them - send every finding to a separate agent on fresh context, because a model recognises its own writing 73.5% of the time and grades it kinder once it does - make that verifier check a real signal like a passing test, never the worker's own word that it finished - shard the fleet across worktrees so parallel workers stop overwriting each other, one rule frozen into every worker: never git stash, never git reset - merge only what came back verified, into one report instead of twenty open chats the catch is the ceiling. at 95% independent work 16 agents return 9.14x rather than the 16 you would guess, and even 256 only reach 18.6x, because the merge and the verify stay serial however wide you fan coordination itself is free plain code and every agent underneath it is billed, so start at twenty files and widen once it works bookmark this, the whole method with all six ready-to-run graphs is written out in the article ↓

this is worth more than most five figure courses 16 claude agents audit an entire repo at once, a second fleet re-checks every finding on fresh context, and the whole thing runs off one diagram instead of a prompt i ran it against my own code and got back 11 endpoints where i never checked who was logged in, 3 of which the verifier threw out before they ever reached me this is Graph Engineering, the layer above prompting, and it runs on the agent you already pay for: - write your plan out, then ask one question at every "and then": does the next step actually read what the previous one produced - the seams that fail that question were never dependencies, so those jobs run at the same time - the arrows that survive are your real edges, and the longest chain of them is your floor that no number of agents shortens - want it faster, cut a false edge instead of adding a worker - fan the independent work out, one agent per item, no shared state between them - send every finding to a separate agent on fresh context, because a model recognises its own writing 73.5% of the time and grades it kinder once it does - make that verifier check a real signal like a passing test, never the worker's own word that it finished - shard the fleet across worktrees so parallel workers stop overwriting each other, one rule frozen into every worker: never git stash, never git reset - merge only what came back verified, into one report instead of twenty open chats the catch is the ceiling. at 95% independent work 16 agents return 9.14x rather than the 16 you would guess, and even 256 only reach 18.6x, because the merge and the verify stay serial however wide you fan coordination itself is free plain code and every agent underneath it is billed, so start at twenty files and widen once it works bookmark this, the whole method with all six ready-to-run graphs is written out in the article ↓

Argona

156,513 Aufrufe • vor 20 Tagen

For 6 months I woke up at 5 AM to catch Asian markets on Polymarket. During that time I lost my girlfriend, gained 8 kg, and got used to drinking coffee instead of breakfast Then I wrote an agent that monitors everything for me while I sleep. In the 1st month income went up 15% and I finally deleted the 5 AM alarm Turns out a half-asleep human trades worse than a 200-line script I thought discipline meant waking up early. In reality it was just stubbornness that cost me money and health. When I finally sat down to build the agent it became clear why Here is what is under the hood: 1. Sentiment analysis powered by Claude. Every 15 minutes the agent runs a feed from 40+ Asian sources: Reuters Asia, Nikkei, South China Morning Post, Yonhap 2. NLP tone classification. It compares sentiment shifts to open markets on Polymarket through the API, and if the news has already dropped but the odds have not reacted yet that is the entry window 3. Kelly criterion. A mathematical formula for position sizing instead of my usual "I will bet more, feeling lucky" 4. A hard stop at 5% of the deposit per trade so that 1 mistake cannot kill the entire account 5. A cooldown between entries so the agent does not stack up a cluster of correlated positions These are exactly the rules I was missing at 5 AM. I knew them perfectly well but consistently ignored them because on adrenaline and caffeine every bet felt like an "obvious opportunity" When I ran a backtest on my old trades it was genuinely painful: 60% of the bets I placed by hand in a half-asleep state would have been rejected by the agent for failing the expected value filter Those were the exact ones dragging the whole result down When I was building my agent I needed a benchmark. A wallet that already trades on similar logic so I could compare my results to someone else's Found 1 that works almost like a mirror of what I described: same Asian markets, same cold calculation without emotion. I still keep it bookmarked and periodically check how it handles the same situations: That is actually the wallet I started with when testing auto-copying through a bot before I launched my own agent. A useful thing if you want to see how a strategy works on someone else's example first and only then build your own:

For 6 months I woke up at 5 AM to catch Asian markets on Polymarket. During that time I lost my girlfriend, gained 8 kg, and got used to drinking coffee instead of breakfast Then I wrote an agent that monitors everything for me while I sleep. In the 1st month income went up 15% and I finally deleted the 5 AM alarm Turns out a half-asleep human trades worse than a 200-line script I thought discipline meant waking up early. In reality it was just stubbornness that cost me money and health. When I finally sat down to build the agent it became clear why Here is what is under the hood: 1. Sentiment analysis powered by Claude. Every 15 minutes the agent runs a feed from 40+ Asian sources: Reuters Asia, Nikkei, South China Morning Post, Yonhap 2. NLP tone classification. It compares sentiment shifts to open markets on Polymarket through the API, and if the news has already dropped but the odds have not reacted yet that is the entry window 3. Kelly criterion. A mathematical formula for position sizing instead of my usual "I will bet more, feeling lucky" 4. A hard stop at 5% of the deposit per trade so that 1 mistake cannot kill the entire account 5. A cooldown between entries so the agent does not stack up a cluster of correlated positions These are exactly the rules I was missing at 5 AM. I knew them perfectly well but consistently ignored them because on adrenaline and caffeine every bet felt like an "obvious opportunity" When I ran a backtest on my old trades it was genuinely painful: 60% of the bets I placed by hand in a half-asleep state would have been rejected by the agent for failing the expected value filter Those were the exact ones dragging the whole result down When I was building my agent I needed a benchmark. A wallet that already trades on similar logic so I could compare my results to someone else's Found 1 that works almost like a mirror of what I described: same Asian markets, same cold calculation without emotion. I still keep it bookmarked and periodically check how it handles the same situations: That is actually the wallet I started with when testing auto-copying through a bot before I launched my own agent. A useful thing if you want to see how a strategy works on someone else's example first and only then build your own:

Blaze

126,874 Aufrufe • vor 4 Monaten

you're paying $20/mo for something your $500 GPU can already do. Gemma 4 26B A4B QAT MoE + Hermes Agent running on a single RTX 4060 (8GB VRAM). Built a vision capable, 100% free, 100% local, private AI assistant that lives in my Chrome browser. No API keys. No cloud. No subscriptions. 100% vibe coded. 0% handholding. It has full context of whatever's on my screen can answer questions, summarize pages, extract data, and see images. Same local model handles everything, no external calls, ever. keep reading for the model and hermes agent tips i learnt while building this locally. Here's the exact setup for anyone running local LLMs on 6-8 GB VRAM: llama.cpp server flags (on my NVIDIA RTX 4060 8gb VRAM): -m gemma-4-26B-A4B-it-qat-UD-Q4_K_XL.gguf --cache-type-k q8_0 --cache-type-v q8_0 -c 150000 --port 8080 Throughput with quantization: Prefill: 200-250 tokens/sec Decode: 20-25 tokens/sec reduce context if oom on 6 gb vram card. Key learnings: - Quantize KV cache to q8 for faster prefill/decode. Prefill goes from 100-150 (unquantized) to 200-250 tok/s (q8). - But watch out, once actual context grows past ~50k tokens on high entropy workloads, q8 KV quantization can cause hallucinations. Low entropy workloads are mostly unaffected. If you see it happening, drop the quantization. This is common across all local models. - In Hermes Agent settings -> Memory & Context, bump compression threshold from default 0.5 to 0.7. Default triggers way too frequent context compression and eats time. Up next: add persistent memory, web search, tool calling, streaming output and whatever you suggest. Running a 26B MoE with vision + 150k context window on 8GB VRAM would've sounded impossible 6 months ago. Works the same on the NVIDIA RTX 3060 Ti, 3070, 4060 Ti, 5060, 2080, or any 8GB card. VRAM is the only requirement. Local AI agents are closer than people think. You just need to know where the knobs are. Model's Unsloth quant hugging face link in the comments. Have you tried Hermes agent by Nous Research yet? What are you building with local LLMs? Drop it below, let's see what this community is shipping.

you're paying $20/mo for something your $500 GPU can already do. Gemma 4 26B A4B QAT MoE + Hermes Agent running on a single RTX 4060 (8GB VRAM). Built a vision capable, 100% free, 100% local, private AI assistant that lives in my Chrome browser. No API keys. No cloud. No subscriptions. 100% vibe coded. 0% handholding. It has full context of whatever's on my screen can answer questions, summarize pages, extract data, and see images. Same local model handles everything, no external calls, ever. keep reading for the model and hermes agent tips i learnt while building this locally. Here's the exact setup for anyone running local LLMs on 6-8 GB VRAM: llama.cpp server flags (on my NVIDIA RTX 4060 8gb VRAM): -m gemma-4-26B-A4B-it-qat-UD-Q4_K_XL.gguf --cache-type-k q8_0 --cache-type-v q8_0 -c 150000 --port 8080 Throughput with quantization: Prefill: 200-250 tokens/sec Decode: 20-25 tokens/sec reduce context if oom on 6 gb vram card. Key learnings: - Quantize KV cache to q8 for faster prefill/decode. Prefill goes from 100-150 (unquantized) to 200-250 tok/s (q8). - But watch out, once actual context grows past ~50k tokens on high entropy workloads, q8 KV quantization can cause hallucinations. Low entropy workloads are mostly unaffected. If you see it happening, drop the quantization. This is common across all local models. - In Hermes Agent settings -> Memory & Context, bump compression threshold from default 0.5 to 0.7. Default triggers way too frequent context compression and eats time. Up next: add persistent memory, web search, tool calling, streaming output and whatever you suggest. Running a 26B MoE with vision + 150k context window on 8GB VRAM would've sounded impossible 6 months ago. Works the same on the NVIDIA RTX 3060 Ti, 3070, 4060 Ti, 5060, 2080, or any 8GB card. VRAM is the only requirement. Local AI agents are closer than people think. You just need to know where the knobs are. Model's Unsloth quant hugging face link in the comments. Have you tried Hermes agent by Nous Research yet? What are you building with local LLMs? Drop it below, let's see what this community is shipping.

Alok

36,031 Aufrufe • vor 1 Monat

Stanford researchers did it again. They just built the agent-native version of Git. When an agent works on a longer task, the run builds up a lot of state. This includes files edited/created, a dev server, a database, installed packages, KV cache, etc. Say the agent is at step 10 and makes a mistake, maybe it misreads a traceback and rewrites a file that was actually fine. The tests start failing, and the run goes off track, although everything through step eight was correct. By default, the agent just tries to fix it, which creates more edits and tool calls. This burns more tokens and grows the context. The other options are a person stepping in to redirect it or restarting the whole run from step one. That's wasteful, because it pays for every model/tool call again and re-prefills the context. Moreover, since an agent's run is non-deterministic, it doesn't reproduce the same early steps anyway. The reason it's hard to just jump back exactly to a previous correct step and resume from there is that the trajectory is only a message log. It records what the agent said and which tools it called, but not the live state underneath. That state includes things like memory, open file handles, child processes, installed packages, /tmp, and KV cache. None of that is in the log. Git can version the files, but it doesn't snapshot the running process or the KV cache. Checking out step eight moves the files back, but the process is still sitting in step-ten memory with a cold cache. Shepherd is a runtime layer by Stanford that records the run as a trace of typed events rather than a flat log. Each agent-environment interaction becomes a commit, similar to Git, but it tracks the live run. Its commit includes the agent process and the filesystem together, copy-on-write, so a branch carries the actual state and not just the files. Going back to a previous step is then a single call that forks from that commit and continues from the exact state. The copy-on-write fork is roughly five times faster than docker commit, and because the prompt prefix through step eight is unchanged, the KV cache is reused over 95% on replay, so early steps aren't reprocessed again. Once the run can be forked, a meta-agent can sit on top and operate it. It watches the trace and reverts as soon as it looks wrong, before the bad write is committed. In practice, it's just Python calling fork, replay, and revert on the trace, rather than a separate control plane wired into the harness. Not everything is reversible though. Files and sandbox changes undo themselves, but a database write has no automatic undo, so it needs a matching undo step set up in advance. Something external, like a sent email or a real charge, can't be undone, so the supervisor's job there is to catch it before it fires. They tested this on a few public benchmarks. On CooperBench, where two agents work on the same codebase, adding a live supervisor took the pair-coding pass rate from 28.8% to 54.7%. It's still early and labeled alpha. The benefit mostly shows up when a run gets branched a lot over a heavy sandbox state, which is exactly where restarting wastes the most tokens and time. If Git was made to make file changes reversible, Shepherd is trying to do the same thing for a live agent run. Shepherd Repo: (don't forget to star it ⭐ ) That said, Shepherd reverts a bad step inside a run. The harness around it, the prompts, tools, and checks the supervisor relies on, still drifts across runs as models and dependencies change. Akshay wrote about making that harness repair itself, where a failing trace gets diagnosed, the fix is verified against the exact input that failed, and the failure is locked as a regression test so it can't recur. Read it below.

Stanford researchers did it again. They just built the agent-native version of Git. When an agent works on a longer task, the run builds up a lot of state. This includes files edited/created, a dev server, a database, installed packages, KV cache, etc. Say the agent is at step 10 and makes a mistake, maybe it misreads a traceback and rewrites a file that was actually fine. The tests start failing, and the run goes off track, although everything through step eight was correct. By default, the agent just tries to fix it, which creates more edits and tool calls. This burns more tokens and grows the context. The other options are a person stepping in to redirect it or restarting the whole run from step one. That's wasteful, because it pays for every model/tool call again and re-prefills the context. Moreover, since an agent's run is non-deterministic, it doesn't reproduce the same early steps anyway. The reason it's hard to just jump back exactly to a previous correct step and resume from there is that the trajectory is only a message log. It records what the agent said and which tools it called, but not the live state underneath. That state includes things like memory, open file handles, child processes, installed packages, /tmp, and KV cache. None of that is in the log. Git can version the files, but it doesn't snapshot the running process or the KV cache. Checking out step eight moves the files back, but the process is still sitting in step-ten memory with a cold cache. Shepherd is a runtime layer by Stanford that records the run as a trace of typed events rather than a flat log. Each agent-environment interaction becomes a commit, similar to Git, but it tracks the live run. Its commit includes the agent process and the filesystem together, copy-on-write, so a branch carries the actual state and not just the files. Going back to a previous step is then a single call that forks from that commit and continues from the exact state. The copy-on-write fork is roughly five times faster than docker commit, and because the prompt prefix through step eight is unchanged, the KV cache is reused over 95% on replay, so early steps aren't reprocessed again. Once the run can be forked, a meta-agent can sit on top and operate it. It watches the trace and reverts as soon as it looks wrong, before the bad write is committed. In practice, it's just Python calling fork, replay, and revert on the trace, rather than a separate control plane wired into the harness. Not everything is reversible though. Files and sandbox changes undo themselves, but a database write has no automatic undo, so it needs a matching undo step set up in advance. Something external, like a sent email or a real charge, can't be undone, so the supervisor's job there is to catch it before it fires. They tested this on a few public benchmarks. On CooperBench, where two agents work on the same codebase, adding a live supervisor took the pair-coding pass rate from 28.8% to 54.7%. It's still early and labeled alpha. The benefit mostly shows up when a run gets branched a lot over a heavy sandbox state, which is exactly where restarting wastes the most tokens and time. If Git was made to make file changes reversible, Shepherd is trying to do the same thing for a live agent run. Shepherd Repo: (don't forget to star it ⭐ ) That said, Shepherd reverts a bad step inside a run. The harness around it, the prompts, tools, and checks the supervisor relies on, still drifts across runs as models and dependencies change. Akshay wrote about making that harness repair itself, where a failing trace gets diagnosed, the fix is verified against the exact input that failed, and the failure is locked as a regression test so it can't recur. Read it below.

Avi Chawla

440,527 Aufrufe • vor 1 Monat

here's how the whole thing works. claude code doesn't care what's behind the API. it just sends requests and expects responses. so i pointed it at my own machine instead of anthropic's servers. llama-server runs the model locally. LiteLLM sits in between and translates the API format. claude code thinks it's talking to claude. it's talking to qwen on localhost. the setup: 2x 3090s, 38 layers on GPU, 10 on CPU. 128K context window. generation is only 7 tok/s but the tradeoff is worth it. 128K means the agent can hold an entire project in memory without losing context midtask. claude code alone loads a 17.5K token system prompt on every request. tool definitions, safety rules, agent behavior. that's your baseline before you even say hello. pushed as far as i could tonight. what surprised me most wasn't the speed. it was the iteration quality. first prompt gave me a working particle sim. second prompt, the model read its own 564 lines, understood the architecture, and added trails, explosions, gravity wells, bloom effects. no handholding. 4bit quantized. 45GB on two consumer cards. running a full coding agent autonomously. detailed article coming. full benchmarks, hardware breakdowns, engine debugging, code quality. everything from setup to what broke and why.

here's how the whole thing works. claude code doesn't care what's behind the API. it just sends requests and expects responses. so i pointed it at my own machine instead of anthropic's servers. llama-server runs the model locally. LiteLLM sits in between and translates the API format. claude code thinks it's talking to claude. it's talking to qwen on localhost. the setup: 2x 3090s, 38 layers on GPU, 10 on CPU. 128K context window. generation is only 7 tok/s but the tradeoff is worth it. 128K means the agent can hold an entire project in memory without losing context midtask. claude code alone loads a 17.5K token system prompt on every request. tool definitions, safety rules, agent behavior. that's your baseline before you even say hello. pushed as far as i could tonight. what surprised me most wasn't the speed. it was the iteration quality. first prompt gave me a working particle sim. second prompt, the model read its own 564 lines, understood the architecture, and added trails, explosions, gravity wells, bloom effects. no handholding. 4bit quantized. 45GB on two consumer cards. running a full coding agent autonomously. detailed article coming. full benchmarks, hardware breakdowns, engine debugging, code quality. everything from setup to what broke and why.

Sudo su

37,623 Aufrufe • vor 5 Monaten

somebody explain this because i refuse to accept it someone ran 48 scored trials and one agent beat a whole fleet of them on all 6 task families, at 0.93 cents a run against 1.9, while openai's best fleet shape was paying $0.008 for every single point of accuracy it bought i read it expecting a hit piece and found the opposite: the fleets that partitioned the dependency graph properly lifted pass rate 14% and cut wall-clock 2.10x on the same tasks, and one of them beat claude code with agent teams the thing that decides it has a name, Graph Engineering, and it is a property of the diagram rather than the model: - partition on the real dependency graph pulled from static analysis, never by folder or by file, because the gains land hardest on the most dependency-dense projects - isolate the structural hub files first, since those are the nodes every partition would otherwise have to share - measure the critical path and treat it as the floor, because a chain that genuinely feeds itself cannot be replaced by more workers and wrapping it in a scheduler does not shorten it - match the topology to the coupling instead of defaulting to parallel: on coupled work a static parallel shape drops below a single agent, so the mismatch is worse than no orchestration - remember each worker serialises its own subtasks, which adds edges inside every agent that were never in your plan - budget the fan-out before you fire it, because three agents already burn roughly three times the tokens and the multiplier compounds across sessions - check worker count against your rate limit, since fifteen workers at ten requests a second walk straight through a hundred-per-second ceiling and cascade - put a script gate in front of the planner: it costs 0.15 seconds and zero tokens, and it lets the expensive model skip 43 to 63% of the steps for at most 1.4 points of accuracy the catch is the coordination tax, and it scales with how clever the shape looks: 58% extra reasoning turns for independent workers, 263% decentralised, 285% centralised, and 515% for the hybrid setup everyone reaches for first the same paper found that hybrid then collapses hardest on tool-heavy work at a 0.452 success rate, while the plainer decentralised shape beat centralised outright despite carrying more overhead, because parallel efficiency is what survives bookmark this, the whole build sits in the article ↓

somebody explain this because i refuse to accept it someone ran 48 scored trials and one agent beat a whole fleet of them on all 6 task families, at 0.93 cents a run against 1.9, while openai's best fleet shape was paying $0.008 for every single point of accuracy it bought i read it expecting a hit piece and found the opposite: the fleets that partitioned the dependency graph properly lifted pass rate 14% and cut wall-clock 2.10x on the same tasks, and one of them beat claude code with agent teams the thing that decides it has a name, Graph Engineering, and it is a property of the diagram rather than the model: - partition on the real dependency graph pulled from static analysis, never by folder or by file, because the gains land hardest on the most dependency-dense projects - isolate the structural hub files first, since those are the nodes every partition would otherwise have to share - measure the critical path and treat it as the floor, because a chain that genuinely feeds itself cannot be replaced by more workers and wrapping it in a scheduler does not shorten it - match the topology to the coupling instead of defaulting to parallel: on coupled work a static parallel shape drops below a single agent, so the mismatch is worse than no orchestration - remember each worker serialises its own subtasks, which adds edges inside every agent that were never in your plan - budget the fan-out before you fire it, because three agents already burn roughly three times the tokens and the multiplier compounds across sessions - check worker count against your rate limit, since fifteen workers at ten requests a second walk straight through a hundred-per-second ceiling and cascade - put a script gate in front of the planner: it costs 0.15 seconds and zero tokens, and it lets the expensive model skip 43 to 63% of the steps for at most 1.4 points of accuracy the catch is the coordination tax, and it scales with how clever the shape looks: 58% extra reasoning turns for independent workers, 263% decentralised, 285% centralised, and 515% for the hybrid setup everyone reaches for first the same paper found that hybrid then collapses hardest on tool-heavy work at a 0.452 success rate, while the plainer decentralised shape beat centralised outright despite carrying more overhead, because parallel efficiency is what survives bookmark this, the whole build sits in the article ↓

Argona

32,433 Aufrufe • vor 16 Tagen

I went a little overboard with Codex last week and burned through my entire weekly allowance in two days. Luckily, my quota reset today. Otherwise, I’m not sure what I would’ve done. It got me thinking: instead of asking one large model to handle everything from start to finish, why not let a stronger model plan the project and review the work, while a model built for execution handles the day-to-day implementation? So I tried it. The result was better than I expected. I used GPT-5.6 Sol in Codex as the decision-maker, then ran Ling-3.0-flash from Ant Ling inside OpenCode as the execution engine. Together, they built a small 3D farming game. Before writing any code, I had Codex create four documents: SPEC.md defined the product scope and the lines we couldn’t cross. ARCHITECTURE.md laid out the isometric coordinate system, state machine, and module boundaries. TASKS.md broke the project into small jobs Ling could tackle one at a time. ACCEPTANCE.md explained how each step would be tested and what “done” actually meant. Then I gave Ling a very straightforward role: You are the execution model for this project. Read all four documents before you begin. Work only on the task assigned for this round. When you’re done, run typecheck, test, and build. If anything fails, read the error, fix it, and run the checks again. Do not move on to the next task early. Ling handled dependency installation, project structure, strict TypeScript configuration, test setup, and a production build in 6 minutes and 3 seconds. It ran into issues with the Vite test config, a TS6310 error, and a missing jsdom dependency along the way. Instead of stopping at the first error, it kept reading the logs and fixing the problems until all three checks passed. The speed was honestly hard to believe. If you exclude the time spent waiting on tools, it was producing more than 100 tokens per second. That made the whole development loop feel noticeably faster. After this experiment, I’m planning to keep using the same workflow. If the task is small, there’s no reason to call an expensive planning model for every single step. If the task is large, handing the entire project to a Flash model in one prompt isn’t a great idea either. The setup that makes more sense to me is: Use a more capable model such as Codex to explore the project, make architectural decisions, and break the work down. Put the constraints into specs, schemas, types, and tests instead of leaving them buried in chat history. Give Ling-3.0-flash a steady stream of clear, verifiable implementation tasks. Report bugs with structured context and actual error logs, rather than saying, “It still doesn’t work.” Bring Codex back in for architecture reviews, visual checks, and changes that affect multiple parts of the project. The point of this setup isn’t to give AI a big “build the whole project” button. It’s to turn software development into a pipeline with a much more sensible cost structure: Codex figures out the plan, sets the boundaries, and catches problems. Ling-3.0-flash moves quickly, calls tools reliably, and works through well-defined tasks at scale. For agent workflows that involve lots of repetitive edits, production tasks, and tool calls, this may be a more practical answer than simply using the biggest model for everything.

I went a little overboard with Codex last week and burned through my entire weekly allowance in two days. Luckily, my quota reset today. Otherwise, I’m not sure what I would’ve done. It got me thinking: instead of asking one large model to handle everything from start to finish, why not let a stronger model plan the project and review the work, while a model built for execution handles the day-to-day implementation? So I tried it. The result was better than I expected. I used GPT-5.6 Sol in Codex as the decision-maker, then ran Ling-3.0-flash from Ant Ling inside OpenCode as the execution engine. Together, they built a small 3D farming game. Before writing any code, I had Codex create four documents: SPEC.md defined the product scope and the lines we couldn’t cross. ARCHITECTURE.md laid out the isometric coordinate system, state machine, and module boundaries. TASKS.md broke the project into small jobs Ling could tackle one at a time. ACCEPTANCE.md explained how each step would be tested and what “done” actually meant. Then I gave Ling a very straightforward role: You are the execution model for this project. Read all four documents before you begin. Work only on the task assigned for this round. When you’re done, run typecheck, test, and build. If anything fails, read the error, fix it, and run the checks again. Do not move on to the next task early. Ling handled dependency installation, project structure, strict TypeScript configuration, test setup, and a production build in 6 minutes and 3 seconds. It ran into issues with the Vite test config, a TS6310 error, and a missing jsdom dependency along the way. Instead of stopping at the first error, it kept reading the logs and fixing the problems until all three checks passed. The speed was honestly hard to believe. If you exclude the time spent waiting on tools, it was producing more than 100 tokens per second. That made the whole development loop feel noticeably faster. After this experiment, I’m planning to keep using the same workflow. If the task is small, there’s no reason to call an expensive planning model for every single step. If the task is large, handing the entire project to a Flash model in one prompt isn’t a great idea either. The setup that makes more sense to me is: Use a more capable model such as Codex to explore the project, make architectural decisions, and break the work down. Put the constraints into specs, schemas, types, and tests instead of leaving them buried in chat history. Give Ling-3.0-flash a steady stream of clear, verifiable implementation tasks. Report bugs with structured context and actual error logs, rather than saying, “It still doesn’t work.” Bring Codex back in for architecture reviews, visual checks, and changes that affect multiple parts of the project. The point of this setup isn’t to give AI a big “build the whole project” button. It’s to turn software development into a pipeline with a much more sensible cost structure: Codex figures out the plan, sets the boundaries, and catches problems. Ling-3.0-flash moves quickly, calls tools reliably, and works through well-defined tasks at scale. For agent workflows that involve lots of repetitive edits, production tasks, and tool calls, this may be a more practical answer than simply using the biggest model for everything.

雪踏乌云

23,107 Aufrufe • vor 20 Tagen

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

1,841,161 Aufrufe • vor 3 Monaten

OpenAI's AgentKit will be so insane, build every step of agents on one platform. These visual agent builders make the whole process of iterating and launching agents far more efficient. It sits on top of the Responses API and unifies the tools that were previously scattered across SDKs and custom orchestration. It lets developers create agent workflows visually, connect data sources securely, and measure performance automatically without coding every layer by hand. The core of AgentKit is the Agent Builder, a drag-and-drop canvas where each node represents an action, guardrail, or decision branch. Developers can link these nodes into multi-agent workflows, preview results instantly, and version each setup. It supports inline evaluation so that developers can see how changes affect output before deploying. The Connector Registry is a single admin panel that manages how data and tools connect across the OpenAI ecosystem. It centralizes integrations like Google Drive, SharePoint, Dropbox, and Microsoft Teams. Large organizations can govern access and flow of data between agents securely under one global console. ChatKit provides a ready-to-use chat interface for embedding agents inside apps or websites. It manages streaming, message threads, and model reasoning displays automatically. Developers can skin the interface to match their product without writing custom front-end code. Under the hood, all these blocks use the same execution core that runs agent reasoning through OpenAI’s APIs. Workflows in Agent Builder compile down to structured instructions for the Responses API, which handles model calls, tool use, and context passing. Connector Registry handles authentication and routing for external tools, while Evals and RFT provide feedback loops that improve agents over time. This integration means developers no longer need to handle orchestration logic, model evaluation pipelines, or safety layers separately. Everything runs natively within OpenAI’s control plane with managed security, automatic versioning, and built-in testing. In short, AgentKit standardizes the entire life cycle of an AI agent—from visual design to deployment and performance tuning—inside a single unified system.

OpenAI's AgentKit will be so insane, build every step of agents on one platform. These visual agent builders make the whole process of iterating and launching agents far more efficient. It sits on top of the Responses API and unifies the tools that were previously scattered across SDKs and custom orchestration. It lets developers create agent workflows visually, connect data sources securely, and measure performance automatically without coding every layer by hand. The core of AgentKit is the Agent Builder, a drag-and-drop canvas where each node represents an action, guardrail, or decision branch. Developers can link these nodes into multi-agent workflows, preview results instantly, and version each setup. It supports inline evaluation so that developers can see how changes affect output before deploying. The Connector Registry is a single admin panel that manages how data and tools connect across the OpenAI ecosystem. It centralizes integrations like Google Drive, SharePoint, Dropbox, and Microsoft Teams. Large organizations can govern access and flow of data between agents securely under one global console. ChatKit provides a ready-to-use chat interface for embedding agents inside apps or websites. It manages streaming, message threads, and model reasoning displays automatically. Developers can skin the interface to match their product without writing custom front-end code. Under the hood, all these blocks use the same execution core that runs agent reasoning through OpenAI’s APIs. Workflows in Agent Builder compile down to structured instructions for the Responses API, which handles model calls, tool use, and context passing. Connector Registry handles authentication and routing for external tools, while Evals and RFT provide feedback loops that improve agents over time. This integration means developers no longer need to handle orchestration logic, model evaluation pipelines, or safety layers separately. Everything runs natively within OpenAI’s control plane with managed security, automatic versioning, and built-in testing. In short, AgentKit standardizes the entire life cycle of an AI agent—from visual design to deployment and performance tuning—inside a single unified system.

Rohan Paul

178,460 Aufrufe • vor 10 Monaten

Claude Cowork Sub-Agents are f*cking cracked 🤯 One prompt → 50 competitor ads analyzed, hooks extracted, and a full creative brief generated. 10 AI agents running in parallel, under 5 minutes. All inside Claude Cowork. Perfect for DTC brands and agencies who are still doing creative research and ad production one task at a time inside Claude. If you're analyzing competitor ads one by one, copying hooks into a spreadsheet manually, writing brief after brief from scratch, and watching Claude's output quality fall off a cliff after the 15th variation because the context window is completely bloated... Sub-agents eliminate the entire bottleneck: → Drop in a spreadsheet of 50 competitor ads and spin up 10 parallel sub-agents → Each sub-agent analyzes 5 ads simultaneously — hooks, angles, CTAs, emotional tone, creative format → They report structured summaries back to the main agent without bloating the context → The main agent synthesizes patterns across all 50 ads into a competitive intel brief → Then spin up another round of sub-agents to generate 30 ad copy variations across 10 personas → Each sub-agent writes for 1-2 personas in a fresh context — so variation 30 is as sharp as variation 1 No analyzing ads one at a time. No context window blowing up halfway through. No copy quality degrading after the first dozen variations. What this gives you: → 50 competitor ads broken down in minutes — hooks, angles, CTAs, formats, all structured → Pattern analysis across the full dataset that you'd miss reviewing ads individually → 30+ ad copy variations with persona-specific messaging that actually stays sharp → A workflow you can save as reusable skills and trigger with one command next time → The same output quality on the last task as the first Built 100% inside Claude Cowork with sub-agents. I put together a full DTC playbook: 5 bulk workflows with copy-paste prompts, the exact sub-agent prompting pattern, batching guidelines, and an honest breakdown of when this setup is worth it vs. when a simpler approach is the better move. Want it for free? > Like this post > Comment "AGENTS" And I'll send it over (must be following so I can DM)

Claude Cowork Sub-Agents are f*cking cracked 🤯 One prompt → 50 competitor ads analyzed, hooks extracted, and a full creative brief generated. 10 AI agents running in parallel, under 5 minutes. All inside Claude Cowork. Perfect for DTC brands and agencies who are still doing creative research and ad production one task at a time inside Claude. If you're analyzing competitor ads one by one, copying hooks into a spreadsheet manually, writing brief after brief from scratch, and watching Claude's output quality fall off a cliff after the 15th variation because the context window is completely bloated... Sub-agents eliminate the entire bottleneck: → Drop in a spreadsheet of 50 competitor ads and spin up 10 parallel sub-agents → Each sub-agent analyzes 5 ads simultaneously — hooks, angles, CTAs, emotional tone, creative format → They report structured summaries back to the main agent without bloating the context → The main agent synthesizes patterns across all 50 ads into a competitive intel brief → Then spin up another round of sub-agents to generate 30 ad copy variations across 10 personas → Each sub-agent writes for 1-2 personas in a fresh context — so variation 30 is as sharp as variation 1 No analyzing ads one at a time. No context window blowing up halfway through. No copy quality degrading after the first dozen variations. What this gives you: → 50 competitor ads broken down in minutes — hooks, angles, CTAs, formats, all structured → Pattern analysis across the full dataset that you'd miss reviewing ads individually → 30+ ad copy variations with persona-specific messaging that actually stays sharp → A workflow you can save as reusable skills and trigger with one command next time → The same output quality on the last task as the first Built 100% inside Claude Cowork with sub-agents. I put together a full DTC playbook: 5 bulk workflows with copy-paste prompts, the exact sub-agent prompting pattern, batching guidelines, and an honest breakdown of when this setup is worth it vs. when a simpler approach is the better move. Want it for free? > Like this post > Comment "AGENTS" And I'll send it over (must be following so I can DM)

Mike Futia

50,124 Aufrufe • vor 5 Monaten

AgentLinter is here! Is your agent sharp & secure? I built AgentLinter, a linter for and agent config files. Here's why. Whether you're vibe-coding or agent-coding, your AI's output quality comes down to one thing: how well you wrote your But managing these files properly? Way harder than it looks. 🎯 The Silent Failure Problem Vague instructions like "write good code" let the agent interpret however it wants. Output gets inconsistent, but nothing throws an error. The failure is silent. Anthropic's own docs say write "Use 2-space indentation" not "Format code properly." But as the file grows, spotting these with your eyes alone is nearly impossible. 🔐 The Security Problem People hard-code API keys and tokens directly into or and commit them, way more often than you'd think. AgentLinter stats show 1 in 5 workspaces has exposed credentials. .gitignore doesn't catch secrets buried inside markdown files. 💥 The Consistency Problem Multiple config files = contradictions. says "be a friendly assistant," says "concise, direct tone." The agent gets confused. references files that don't exist. Past 5 files, these conflicts triple. So I thought: is code. Code has ESLint. Why doesn't this have a linter? 🔍 What AgentLinter Does It diagnoses your agent config across 8 categories: 1) Structure: file organization 2) Clarity: instruction specificity 3) Completeness: missing definitions 4) Security: exposed secrets 5) Consistency: cross-file contradictions 6) Memory: session handoff 7) Runtime Config: gateway/auth settings 8) Skill Safety: dangerous shell commands & injection patterns Each scored 0–100 with concrete fix suggestions. Write "be helpful" and it tells you to specify response length, tone, and format. Find an API key? Instant CRITICAL alert to rotate. 🔒 Privacy-First & 100% Local Everything runs on your machine. Files never leave. Only the results are shared, and you can turn that off in settings. This matters — these files can contain system prompts, security rules, and personal context. Fully open source, MIT license, 100% free. 🛠️ Multi-Tool Support Works with Claude Code, Cursor, Windsurf, and Clawdbot. Detects for project mode, or clawdbot.json for agent mode and adjusts diagnostics automatically. 🚀 Get Started with one line npx agentlinter Node.js 18+, no config needed. Run it, check your score, fix what needs fixing. Happy vibe-coding & happy agent life! 🤙 Website: Github:

AgentLinter is here! Is your agent sharp & secure? I built AgentLinter, a linter for and agent config files. Here's why. Whether you're vibe-coding or agent-coding, your AI's output quality comes down to one thing: how well you wrote your But managing these files properly? Way harder than it looks. 🎯 The Silent Failure Problem Vague instructions like "write good code" let the agent interpret however it wants. Output gets inconsistent, but nothing throws an error. The failure is silent. Anthropic's own docs say write "Use 2-space indentation" not "Format code properly." But as the file grows, spotting these with your eyes alone is nearly impossible. 🔐 The Security Problem People hard-code API keys and tokens directly into or and commit them, way more often than you'd think. AgentLinter stats show 1 in 5 workspaces has exposed credentials. .gitignore doesn't catch secrets buried inside markdown files. 💥 The Consistency Problem Multiple config files = contradictions. says "be a friendly assistant," says "concise, direct tone." The agent gets confused. references files that don't exist. Past 5 files, these conflicts triple. So I thought: is code. Code has ESLint. Why doesn't this have a linter? 🔍 What AgentLinter Does It diagnoses your agent config across 8 categories: 1) Structure: file organization 2) Clarity: instruction specificity 3) Completeness: missing definitions 4) Security: exposed secrets 5) Consistency: cross-file contradictions 6) Memory: session handoff 7) Runtime Config: gateway/auth settings 8) Skill Safety: dangerous shell commands & injection patterns Each scored 0–100 with concrete fix suggestions. Write "be helpful" and it tells you to specify response length, tone, and format. Find an API key? Instant CRITICAL alert to rotate. 🔒 Privacy-First & 100% Local Everything runs on your machine. Files never leave. Only the results are shared, and you can turn that off in settings. This matters — these files can contain system prompts, security rules, and personal context. Fully open source, MIT license, 100% free. 🛠️ Multi-Tool Support Works with Claude Code, Cursor, Windsurf, and Clawdbot. Detects for project mode, or clawdbot.json for agent mode and adjusts diagnostics automatically. 🚀 Get Started with one line npx agentlinter Node.js 18+, no config needed. Run it, check your score, fix what needs fixing. Happy vibe-coding & happy agent life! 🤙 Website: Github:

Simon Kim

44,224 Aufrufe • vor 6 Monaten

my 8 GB VRAM gaming laptop is absolutely going to hate me for this. but I still did it. ran a 31b dense model (Gemma 4 31b Q4) with only 8 GB VRAM last week I ran Gemma 4 26B A4B a mixture of experts model on my RTX 4060 and hit 25–28 tokens/sec using llama.cpp's new MTP support. smooth. snappy. but MoE has a secret: it only activates 4B parameters per token despite having 26B total. that's why it flies. so the real question started haunting me. what if I throw a full, no tricks, every parameter fires on every token, 31B DENSE model at the same machine? # Hardware: GPU: NVIDIA RTX 4060, 8 GB VRAM RAM: 16 GB CPU: Intel Core i7 H Laptop. Gaming. Modest. The model: gemma-4-31B-it-qat-UD-Q4_K_XL.gguf (model's unsloth huggingface link in the comments) This is Google DeepMind's flagship dense model in the Gemma 4 family that can run on single consumer GPU. It packs a hybrid attention architecture, supports up to 256K context natively, and is QAT (Quantization Aware Training) optimized, meaning it retains far more quality than standard post training quants at the same bit depth. This is NOT the MoE. This is 31 BILLION dense parameters, every single one of them loaded. # the flags I used: -m gemma-4-31B-it-qat-UD-Q4_K_XL.gguf -cnv --spec-type draft-mtp --spec-draft-model mtp-gemma-4-31B-it.gguf --spec-draft-n-max 8 --spec-draft-p-min 0.6 -c 6000 -v Multi Token Prediction (MTP) is still active here. Separate draft GGUF required, same as the 26B setup. # Results: → Decode: ~3 tokens/sec → Prefill: ~2 tokens/sec → Context: 6000 tokens → Hardware crying quietly in the corner: yes so is 3 tps actually usable? For real time back and forth chat? Not ideal. You're not having a fluid conversation at 3 tps. but slow ≠ useless. And this is where it gets genuinely interesting. think about how senior devs actually work in a real team. But when something is architectural, deeply complex, or needs serious reasoning? they walk down the hall and escalate to the senior. That's exactly the local AI agent architecture this unlocks: → Fast orchestrator model (Gemma 4 26B MoE at 25+ tps) handles routing, simple queries, tool calls, memory. The junior dev. → Gemma 4 31B dense is the senior, called only when the fast model genuinely hits a wall. Hard multi step reasoning. Complex code generation. Deep architectural decisions. The agentic loop stays fast. Only the hard hops touch the 31B. That's a legitimate production grade local AI architecture on a budget hardware. (requires 2 8gb gpus) other workflows where 3 tps is completely fine: - overnight batch jobs. summarize documents, extract structured data, review code. Fire it off. Sleep. wake up to results. - One shot deep reasoning - Silent code audit loops, you write and test, the 31B reviews diffs and flags issues in the background between your sprints - Any workflow where output quality > output speed A few weeks ago, nobody was running a 30B+ dense model on a single consumer GPU with 8 GB VRAM. At all. Now we're doing it on an Intel i7-H gaming laptop with a NVIDIA RTX 4060, thanks to llama.cpp + QAT quants + MTP speculative drafting. Google DeepMind said the Gemma 4 31B targets "consumer GPUs and workstations." They were not exaggerating. The hardware bar to run serious frontier class models locally keeps dropping. the tools are here. the models are here. you just have to be willing to abuse your laptop a little. what workflows would you actually run on a local 3 tps 31B dense model? genuinely curious. drop it below.

my 8 GB VRAM gaming laptop is absolutely going to hate me for this. but I still did it. ran a 31b dense model (Gemma 4 31b Q4) with only 8 GB VRAM last week I ran Gemma 4 26B A4B a mixture of experts model on my RTX 4060 and hit 25–28 tokens/sec using llama.cpp's new MTP support. smooth. snappy. but MoE has a secret: it only activates 4B parameters per token despite having 26B total. that's why it flies. so the real question started haunting me. what if I throw a full, no tricks, every parameter fires on every token, 31B DENSE model at the same machine? # Hardware: GPU: NVIDIA RTX 4060, 8 GB VRAM RAM: 16 GB CPU: Intel Core i7 H Laptop. Gaming. Modest. The model: gemma-4-31B-it-qat-UD-Q4_K_XL.gguf (model's unsloth huggingface link in the comments) This is Google DeepMind's flagship dense model in the Gemma 4 family that can run on single consumer GPU. It packs a hybrid attention architecture, supports up to 256K context natively, and is QAT (Quantization Aware Training) optimized, meaning it retains far more quality than standard post training quants at the same bit depth. This is NOT the MoE. This is 31 BILLION dense parameters, every single one of them loaded. # the flags I used: -m gemma-4-31B-it-qat-UD-Q4_K_XL.gguf -cnv --spec-type draft-mtp --spec-draft-model mtp-gemma-4-31B-it.gguf --spec-draft-n-max 8 --spec-draft-p-min 0.6 -c 6000 -v Multi Token Prediction (MTP) is still active here. Separate draft GGUF required, same as the 26B setup. # Results: → Decode: ~3 tokens/sec → Prefill: ~2 tokens/sec → Context: 6000 tokens → Hardware crying quietly in the corner: yes so is 3 tps actually usable? For real time back and forth chat? Not ideal. You're not having a fluid conversation at 3 tps. but slow ≠ useless. And this is where it gets genuinely interesting. think about how senior devs actually work in a real team. But when something is architectural, deeply complex, or needs serious reasoning? they walk down the hall and escalate to the senior. That's exactly the local AI agent architecture this unlocks: → Fast orchestrator model (Gemma 4 26B MoE at 25+ tps) handles routing, simple queries, tool calls, memory. The junior dev. → Gemma 4 31B dense is the senior, called only when the fast model genuinely hits a wall. Hard multi step reasoning. Complex code generation. Deep architectural decisions. The agentic loop stays fast. Only the hard hops touch the 31B. That's a legitimate production grade local AI architecture on a budget hardware. (requires 2 8gb gpus) other workflows where 3 tps is completely fine: - overnight batch jobs. summarize documents, extract structured data, review code. Fire it off. Sleep. wake up to results. - One shot deep reasoning - Silent code audit loops, you write and test, the 31B reviews diffs and flags issues in the background between your sprints - Any workflow where output quality > output speed A few weeks ago, nobody was running a 30B+ dense model on a single consumer GPU with 8 GB VRAM. At all. Now we're doing it on an Intel i7-H gaming laptop with a NVIDIA RTX 4060, thanks to llama.cpp + QAT quants + MTP speculative drafting. Google DeepMind said the Gemma 4 31B targets "consumer GPUs and workstations." They were not exaggerating. The hardware bar to run serious frontier class models locally keeps dropping. the tools are here. the models are here. you just have to be willing to abuse your laptop a little. what workflows would you actually run on a local 3 tps 31B dense model? genuinely curious. drop it below.

Alok

63,689 Aufrufe • vor 1 Monat