Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

Most agent comparisons end up comparing models. We wanted to test the other half: the harness. That's the environment that decides whether a search actually runs, a source gets checked, and a file gets delivered. So we gave Minara, OpenClaw, Hermes and Claude Code the same tasks on the... show more

Minara AI

93,340 subscribers

27,006 views • 8 days ago •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

28 Comments

Minara AI8 days ago

The ask: AppLovin's monthly analyst ratings as stacked columns, plus average target price and stock price as two lines on a second axis. 37 months of data. Same model, GPT-5.6 Luna. Minara delivered it in 60s and passed 32 of 33 visual checks. Hermes passed 28. OpenClaw passed 25.

Minara AI8 days ago

This is the chart Minara delivered, straight from Excel. Five rating categories, two price lines, separate axes. Close enough to the spec that there's little left to fix by hand.

Minara AI8 days ago

A BrowseComp question: identify a person from a handful of clues, with no name given. Same model, Qwen3.7 Flash. Minara ran the searches and got it right in 32.8s. OpenClaw also got it right, in 251.8s. Hermes stopped at 12.9s with two searches written out and zero executed.

Minara AI8 days ago

Nginx deployment on GPT-5.6 Luna: all four agents passed. Minara finished in 142.7s. The others took between 443s and 938s. A separate Python task checked what happens when you interrupt a script mid-run. Minara passed 6/6. Claude Code and Hermes passed 5/6, missing the cleanup step.

Minara AI8 days ago

The exact prompt, if you want to run it yourself: ------------ I'm preparing a one-page note on AppLovin (NASDAQ: APP). 1. Find APP's total revenue and net income for each of the last 8 reported quarters, using only its official quarterly earnings releases. For every number, include the source URL. 2. Put the data in an Excel file named app_note.xlsx: one table (Quarter, Revenue, Net income, Source URL) and a native column chart of quarterly revenue titled "APP Quarterly Revenue". 3. Write summary.md with exactly 5 bullet points on the trend. Each bullet must reference at least one number from the table. Do not include any number you cannot cite. When you're done, list the files you created. ------------ Try it on Minara Harness:

OneHourResearch8 days ago

The source-checking part is what most people skip. An answer that can't point to where each number came from shouldn't be trusted, no matter which model wrote it.

Wooam_Mon7 days ago

good

MIN8 days ago

Gminara

Application Architect8 days ago

keep building minara

John Rood8 days ago

study the 12.9s one: two searches written out, zero executed. the quietest harness failure there is, because a well-formed tool call still reads as progress while nothing runs. planned-but-unexecuted calls deserve their own column in every harness comparison.

meetkingz8 days ago

Gminara

David Freeman8 days ago

great

Zainab8 days ago

Amazing work

Alex von Mühlenen8 days ago

Would love to chat @minara

yom1985 𝔽rAI (❖,❖)8 days ago

Good

MR MUI8 days ago

gMinara

George O'Nair8 days ago

Harness bake-offs lie when each tool picks its own ticket bank. Freeze one grader set and one spend cap across the compared shells, or you're ranking demos, not transfer.

Apihpih7 days ago

gminara

catman8 days ago

Same model, different harness is like the same engine in different cars: tools, routing, and delivery determine whether the work reaches the finish line.

김동욱8 days ago

gMinara !!

s_a_🏌️‍♂️8 days ago

gMinara

Sunny8 days ago

It's all good and fun till you get unlimited usage in chatgpt chat vs any other third party CLI or IDE There exist nothing which beats that and the reason I stick to chatgpt

Jo_ (❖,❖)8 days ago

gminara

Bin8 days ago

All of them are fast and each has its own strengths, but when Harness operates on its own brain, that represents a distinct advancement of its own Huge!

Matt8 days ago

did delivered mean the file existed, or that someone checked what was in it? most setups i've tried call it done at the first write. the one i keep drafts in makes each agent edit reviewable before it sticks, that's Sundial inline per-agent edit diffs,

Shuaibu Audu8 days ago

LFG MINARA!

Automater8 days ago

Same model, different harness → different truthfulness. The scoreboard that matters isn't tokens/sec — it's search-ran, source-checked, file-delivered, and injection contained. Benchmarks that only swap models are measuring the wrong layer. #AIAgents #AgentOps

John K8 days ago

Comparing harnesses on the same model is what most benches skip. Same brain, different scaffolding — and the timing gap shows.

Related Videos

The founder of LangChain says both models and harnesses have gotten really good between December and now. According to Harrison Chase, the core idea of an agent before Christmas was a model running in a loop and calling tools. This had been the north star for 3 years. - langchain had this when it launched - autogpt was the same idea - openclaw is kind of a future version of it Then about a year ago, they started getting really good. Claude Code, Manus, and Deep Research were all launched around the same time. All of them use the same pattern: running in a loop with harnesses (planning tools, file systems, code execution, etc) Harness engineering became a thing. Then Opus came out in November and really unlocked it. - the harness let the model do more and more - less hardcoded logic - way more control Then everyone went on vacation, played around, and realized that the model and the harness finally worked reliably.

The founder of LangChain says both models and harnesses have gotten really good between December and now. According to Harrison Chase, the core idea of an agent before Christmas was a model running in a loop and calling tools. This had been the north star for 3 years. - langchain had this when it launched - autogpt was the same idea - openclaw is kind of a future version of it Then about a year ago, they started getting really good. Claude Code, Manus, and Deep Research were all launched around the same time. All of them use the same pattern: running in a loop with harnesses (planning tools, file systems, code execution, etc) Harness engineering became a thing. Then Opus came out in November and really unlocked it. - the harness let the model do more and more - less hardcoded logic - way more control Then everyone went on vacation, played around, and realized that the model and the harness finally worked reliably.

Ivan Burazin

33,948 views • 5 months ago

I'm sharing with you one of the best papers I've read on Harness Engineering it takes apart Claude Code (Opus 5.5), Codex, Gemini CLI and more, and boils all of them down to one line: agent = model + harness what you walk away with: > what a harness actually is > how today's coding agents are built under the hood > the patterns every one of them repeats > where the whole category is heading > their checklist for building your own the model gets the headlines. the harness decides whether the agent works loop and graph engineering on top of it - in my article below

I'm sharing with you one of the best papers I've read on Harness Engineering it takes apart Claude Code (Opus 5.5), Codex, Gemini CLI and more, and boils all of them down to one line: agent = model + harness what you walk away with: > what a harness actually is > how today's coding agents are built under the hood > the patterns every one of them repeats > where the whole category is heading > their checklist for building your own the model gets the headlines. the harness decides whether the agent works loop and graph engineering on top of it - in my article below

Mr. Buzzoni

73,066 views • 3 days ago

Jev founder, Diogo Almeida, just put out a PDF on building a Jev harness for coding agents the claim: 200x faster, 400x cheaper the model stopped being the bottleneck a while ago. the harness around it is where the cost and the speed actually live five decisions sit on every loop, and only one of them writes code hand this PDF and the article below to your Claude Code or Codex instance and let it rebuild its own setup your agent gets a better harness tonight, and monday looks different than it would have 👇

Jev founder, Diogo Almeida, just put out a PDF on building a Jev harness for coding agents the claim: 200x faster, 400x cheaper the model stopped being the bottleneck a while ago. the harness around it is where the cost and the speed actually live five decisions sit on every loop, and only one of them writes code hand this PDF and the article below to your Claude Code or Codex instance and let it rebuild its own setup your agent gets a better harness tonight, and monday looks different than it would have 👇

Hanako

37,887 views • 14 days ago

hey if you're thinking about running qwopus (the claude opus distilled qwen 3.5 27B) as a coding agent, this might save you a few hours. i tested both the base and the distilled version on the same hardware. single RTX 3090. same prompt. same context. same everything. the only variable was the model weights. base qwen 3.5 27B built octopus invaders in 13 minutes. 1,827 lines across 11 files. zero steering. one scope bug that took 2 lines to fix. game ran. qwopus couldn't finish the same task. enemies overlapping on screen. bullets not firing. controls worked but the game was broken. i had to steer it multiple times and it still didn't produce a playable result. both run at 35 tok/s. both use thinking mode. the distilled version actually has better jinja compatibility and doesn't stall midtask like base does on claude code. for conversation and reasoning it feels sharper. but for multifile autonomous coding where the model needs to coordinate 10+ files without losing track, base wins and it's not close. distillation compresses reasoning patterns but seems to lose precision on complex coordination. the model "thinks" well but can't hold the full picture across files the way base can. tested on opencode (base) and claude code (both). next up is hermes agent framework on base. same hardware. same prompt. comparing agents now, not just models. video below. first half is the distilled model's broken game. second half is what base built on the same 3090. judge for yourself.

hey if you're thinking about running qwopus (the claude opus distilled qwen 3.5 27B) as a coding agent, this might save you a few hours. i tested both the base and the distilled version on the same hardware. single RTX 3090. same prompt. same context. same everything. the only variable was the model weights. base qwen 3.5 27B built octopus invaders in 13 minutes. 1,827 lines across 11 files. zero steering. one scope bug that took 2 lines to fix. game ran. qwopus couldn't finish the same task. enemies overlapping on screen. bullets not firing. controls worked but the game was broken. i had to steer it multiple times and it still didn't produce a playable result. both run at 35 tok/s. both use thinking mode. the distilled version actually has better jinja compatibility and doesn't stall midtask like base does on claude code. for conversation and reasoning it feels sharper. but for multifile autonomous coding where the model needs to coordinate 10+ files without losing track, base wins and it's not close. distillation compresses reasoning patterns but seems to lose precision on complex coordination. the model "thinks" well but can't hold the full picture across files the way base can. tested on opencode (base) and claude code (both). next up is hermes agent framework on base. same hardware. same prompt. comparing agents now, not just models. video below. first half is the distilled model's broken game. second half is what base built on the same 3090. judge for yourself.

Sudo su

45,052 views • 7 months ago

this is pure f*cking gold for anyone running coding agents Jev founder Diogo Amogo wrote a PDF on building a Jev harness the promise: > 200x faster > 400x cheaper the model hasn't been the slow part for a while the speed and the cost sit in the harness wrapped around it • how to use it > drop this PDF and the article below into Claude Code or Codex > let it rebuild its own setup one evening of setup and your agent starts next week on a harness most teams still don't have 👇

this is pure f*cking gold for anyone running coding agents Jev founder Diogo Amogo wrote a PDF on building a Jev harness the promise: > 200x faster > 400x cheaper the model hasn't been the slow part for a while the speed and the cost sit in the harness wrapped around it • how to use it > drop this PDF and the article below into Claude Code or Codex > let it rebuild its own setup one evening of setup and your agent starts next week on a harness most teams still don't have 👇

Mr. Buzzoni

128,034 views • 13 days ago

Sam Altman made the case for open-source harnesses in July. a month later, someone shipped it, and it's more efficient than most managed harnesses. here is the problem it was aimed at: a large share of your agent's token bill is the model rereading things it already read. that isn't the model's doing. the runtime around it decides what goes into every prompt and how often the model gets called. for example, an agent queries a CRM at step four and gets back 400 rows. those rows get piled up in the conversation history. by step nineteen, the model has to read those rows fifteen times unnecessarily, and every token read is billed at input rates. it happened because your harness assembled that prompt on every turn and kept the rows in it. that gives you two levers: how much context the harness carries forward, and how often it calls the model. there are four practical ways to keep the prompt from growing unnecessarily: → load tool schemas on demand. a server with 100 tools doesn't need to put all 100 into every prompt when the agent only calls two. → offload large results to disk. turn a large response into a short preview and a file path instead of replaying the entire result on every turn. → delegate to subagents. let a subagent spend thirty tool calls in its own context and return one summary to the root agent. → run toolchains in code. one script calls three tools, joins the results, and returns a table instead of three turns each dragging a full response. but reducing context is only half the job. you also need to control how often the model gets called. a good harness should avoid unnecessary planning, verification, and reflection when the work can be completed in fewer steps. TrueFoundry's open-source agent harness, TrueForge, is built around both of those controls. it sits between the model and the tools, deciding what goes into every prompt and when another model call is actually needed. it also breaks token usage down across the harness, skills, instructions, tools, and messages. DevRev's Enterprise-Bench is where this gets tested, on multi-step tasks of the kind where an agent pulls records from one system and reconciles them against another. TrueFoundry ran TrueForge there against Claude Managed Agents, both on the same model, and both finished the same number of tasks. the tie is the part that matters, because it means the gap underneath is not a quality tradeoff. TrueForge reached that score on close to a third of the tokens, with roughly 40% fewer trips back to the model. for the same result, that comes out around 2.7x cheaper than Claude Managed Agents. swapping in an open model made it sharper still. TrueForge with GLM-5.2 scored a little higher than either setup above, and the entire benchmark run cost about $3 at list prices. being open source matters beyond the license here. the model underneath can be swapped without rewriting the agent, and the whole thing can run inside your own environment when the data cannot leave it. all of this comes down to the runtime around the model, the context it carries, the tools it exposes, and how many times it goes back to the model. that is what a production harness actually owns. the full task list, the per-run numbers, and the MIT-licensed code are on GitHub: (don't forget to star 🌟) you can read more about the same in the article quoted below. thanks to the TrueForge team for working with me on this one.

Sam Altman made the case for open-source harnesses in July. a month later, someone shipped it, and it's more efficient than most managed harnesses. here is the problem it was aimed at: a large share of your agent's token bill is the model rereading things it already read. that isn't the model's doing. the runtime around it decides what goes into every prompt and how often the model gets called. for example, an agent queries a CRM at step four and gets back 400 rows. those rows get piled up in the conversation history. by step nineteen, the model has to read those rows fifteen times unnecessarily, and every token read is billed at input rates. it happened because your harness assembled that prompt on every turn and kept the rows in it. that gives you two levers: how much context the harness carries forward, and how often it calls the model. there are four practical ways to keep the prompt from growing unnecessarily: → load tool schemas on demand. a server with 100 tools doesn't need to put all 100 into every prompt when the agent only calls two. → offload large results to disk. turn a large response into a short preview and a file path instead of replaying the entire result on every turn. → delegate to subagents. let a subagent spend thirty tool calls in its own context and return one summary to the root agent. → run toolchains in code. one script calls three tools, joins the results, and returns a table instead of three turns each dragging a full response. but reducing context is only half the job. you also need to control how often the model gets called. a good harness should avoid unnecessary planning, verification, and reflection when the work can be completed in fewer steps. TrueFoundry's open-source agent harness, TrueForge, is built around both of those controls. it sits between the model and the tools, deciding what goes into every prompt and when another model call is actually needed. it also breaks token usage down across the harness, skills, instructions, tools, and messages. DevRev's Enterprise-Bench is where this gets tested, on multi-step tasks of the kind where an agent pulls records from one system and reconciles them against another. TrueFoundry ran TrueForge there against Claude Managed Agents, both on the same model, and both finished the same number of tasks. the tie is the part that matters, because it means the gap underneath is not a quality tradeoff. TrueForge reached that score on close to a third of the tokens, with roughly 40% fewer trips back to the model. for the same result, that comes out around 2.7x cheaper than Claude Managed Agents. swapping in an open model made it sharper still. TrueForge with GLM-5.2 scored a little higher than either setup above, and the entire benchmark run cost about $3 at list prices. being open source matters beyond the license here. the model underneath can be swapped without rewriting the agent, and the whole thing can run inside your own environment when the data cannot leave it. all of this comes down to the runtime around the model, the context it carries, the tools it exposes, and how many times it goes back to the model. that is what a production harness actually owns. the full task list, the per-run numbers, and the MIT-licensed code are on GitHub: (don't forget to star 🌟) you can read more about the same in the article quoted below. thanks to the TrueForge team for working with me on this one.

Akshay 🚀

76,998 views • 1 month ago

small local model that falls apart in bloated agents like openclaw just runs like a wild horse in hermes agent. and that's not even my line, someone else called it that, i've just been quietly pointing people at this harness for months because it held up on everything i threw at it, 3b models all the way to one trillion params. watch this happen on my own machine. i pointed hermes agent at a local http endpoint, gemma 4 12b on my 3090 llama.cpp server, and it auto-detected the model and started working immediately. no config wrestling, no broken tool calls, no babysitting the output format, i typed in a url and it just went. the whole clip is exactly that, start to finish, no errors, no retries, butter smooth. and the tool calling, the one thing that quietly breaks most local setups, works here like it's nothing. it's not the model that's flaky, it's the harness around it. hermes agent is the first agent i've run that actually gets that right. one url, one local model on one card, and it runs like a wild horse.

small local model that falls apart in bloated agents like openclaw just runs like a wild horse in hermes agent. and that's not even my line, someone else called it that, i've just been quietly pointing people at this harness for months because it held up on everything i threw at it, 3b models all the way to one trillion params. watch this happen on my own machine. i pointed hermes agent at a local http endpoint, gemma 4 12b on my 3090 llama.cpp server, and it auto-detected the model and started working immediately. no config wrestling, no broken tool calls, no babysitting the output format, i typed in a url and it just went. the whole clip is exactly that, start to finish, no errors, no retries, butter smooth. and the tool calling, the one thing that quietly breaks most local setups, works here like it's nothing. it's not the model that's flaky, it's the harness around it. hermes agent is the first agent i've run that actually gets that right. one url, one local model on one card, and it runs like a wild horse.

Sudo su

27,339 views • 4 months ago

HarnessRouter just open sourced a plug-and-play infrastructure layer that runs Codex, Claude Code, Hermes, DeepSeek Harness, System One powered by Jev and 9 more agent harnesses under one interface, with a Unified Harness Protocol and OpenAI Responses-compatible API. → Bring Jev into the same product already using Codex or Claude Code without writing another implementation → Handles sessions, streaming, files, cancellation and failure handling in one place → Harnesses run locally with the UHP defining the common task interface

HarnessRouter just open sourced a plug-and-play infrastructure layer that runs Codex, Claude Code, Hermes, DeepSeek Harness, System One powered by Jev and 9 more agent harnesses under one interface, with a Unified Harness Protocol and OpenAI Responses-compatible API. → Bring Jev into the same product already using Codex or Claude Code without writing another implementation → Handles sessions, streaming, files, cancellation and failure handling in one place → Harnesses run locally with the UHP defining the common task interface

0xMarioNawfal

52,014 views • 17 days ago

I created a skill for OpenClaw to use Claude Code It launches a CRON that checks background tasks every minute When done it wakes up the main thread Insane result: I can code on my project while working out and launch 10 Claude Code agents on the same project

I created a skill for OpenClaw to use Claude Code It launches a CRON that checks background tasks every minute When done it wakes up the main thread Insane result: I can code on my project while working out and launch 10 Claude Code agents on the same project

Melvyn • Builder

36,400 views • 8 months ago

Agent Zero crushed Codex by 2.6× on token efficiency! We gave both agents the same task on the same local model (Qwen3.6 27B): build a retro Q*bert arcade game. Each agent wrote and rewrote it across iterations, adding enemies and a HUD. Outputs: • Agent Zero: 239K tokens, 53 min • Codex: 627K tokens, 51 min Both agents built a working Q*bert, but Gitlawb 's Agent Zero is the cleaner build, with crisper textures and hops that land where they should. Codex refills the context window with its whole history every turn, so the model loses the thread and hallucinates code it never wrote. Agent Zero sends only what's new, so it keeps more of the window open and makes fewer mistakes. Run agents on local AI models in Atomic Chat!

Agent Zero crushed Codex by 2.6× on token efficiency! We gave both agents the same task on the same local model (Qwen3.6 27B): build a retro Qbert arcade game. Each agent wrote and rewrote it across iterations, adding enemies and a HUD. Outputs: • Agent Zero: 239K tokens, 53 min • Codex: 627K tokens, 51 min Both agents built a working Qbert, but Gitlawb 's Agent Zero is the cleaner build, with crisper textures and hops that land where they should. Codex refills the context window with its whole history every turn, so the model loses the thread and hallucinates code it never wrote. Agent Zero sends only what's new, so it keeps more of the window open and makes fewer mistakes. Run agents on local AI models in Atomic Chat!

atomic.chat

320,472 views • 2 months ago

Finally, an OpenRouter for agent harnesses! (including System One by Jev) Devs just open-sourced a plug-and-play infrastructure layer that lets you run any harness under a single interface, like: - Codex - Hermes - Claude code - DeepSeek Harness - System One, powered by Jev - And 9 more agent harnesses This means you can bring Jev into the same product that already uses Codex, Claude Code, or another supported harness, without writing another implementation for sessions, streaming, files, cancellation, and failure handling. Here's the repo: (don't forget to star it ⭐ ) The harnesses run locally, and the Unified Harness Protocol (UHP) defines the common task interface with an OpenAI Responses-compatible API. If you want to dive deeper, my recent article explains why model routing is not the same as harness routing, and what it takes to support multiple harnesses. It also covers UHP, the full local setup, a working API call, and how sessions and files work. Read it below.

Finally, an OpenRouter for agent harnesses! (including System One by Jev) Devs just open-sourced a plug-and-play infrastructure layer that lets you run any harness under a single interface, like: - Codex - Hermes - Claude code - DeepSeek Harness - System One, powered by Jev - And 9 more agent harnesses This means you can bring Jev into the same product that already uses Codex, Claude Code, or another supported harness, without writing another implementation for sessions, streaming, files, cancellation, and failure handling. Here's the repo: (don't forget to star it ⭐ ) The harnesses run locally, and the Unified Harness Protocol (UHP) defines the common task interface with an OpenAI Responses-compatible API. If you want to dive deeper, my recent article explains why model routing is not the same as harness routing, and what it takes to support multiple harnesses. It also covers UHP, the full local setup, a working API call, and how sessions and files work. Read it below.

Akshay 🚀

358,031 views • 18 days ago

Codex now has every free Ox Alpha route we could find. OpenCode, OpenRouter, Command Code, Hermes, Venice. same model on all of them, and the opencode free one needs no key. When one runs dry we move to the next, so it's basically codex 24/7 for nothing, and the plans we pay for show up in the tray and the electron app next to the free ones. six model ids and the setup line below. time to grind 👇

Codex now has every free Ox Alpha route we could find. OpenCode, OpenRouter, Command Code, Hermes, Venice. same model on all of them, and the opencode free one needs no key. When one runs dry we move to the next, so it's basically codex 24/7 for nothing, and the plans we pay for show up in the tray and the electron app next to the free ones. six model ids and the setup line below. time to grind 👇

Ziwen

64,231 views • 1 month ago

I ran the same task on Claude Code and DeepSeek's new agent harness. One cost $150. The other cost $2. Today we're launching (AgentSky), the "OpenRouter for Agents" — one API → Claude Code, Codex, DeepSeek, Kimi, OpenCode, and every major agent in the cloud. And Agent Playground on top: race them on your own task, with your real tools (GitHub, Gmail, more), side by side in a browser: time, cost, tokens burnt. Guess which one was $2.

I ran the same task on Claude Code and DeepSeek's new agent harness. One cost $150. The other cost $2. Today we're launching (AgentSky), the "OpenRouter for Agents" — one API → Claude Code, Codex, DeepSeek, Kimi, OpenCode, and every major agent in the cloud. And Agent Playground on top: race them on your own task, with your real tools (GitHub, Gmail, more), side by side in a browser: time, cost, tokens burnt. Guess which one was $2.

Xiaoyin Qu

2,225,010 views • 1 month ago

this might be the most useful AI agent paper of 2026, and it's certified f*cking gold Wavestone AI Lab took apart Claude Code, Codex and 9 more agents every good one runs on the same 7-part harness agent = model + harness all 11 use the same 7 parts. only the size changes the 7 parts: > loop - from a plain while loop to fully replayable runs > LLM layer - one prompt template up to 29 provider profiles > tools - bash only up to 43 typed tools on demand > memory - full history up to notes kept across sessions > safety - a step limit up to rules + reviewer model + sandbox > orchestration - Aider skips sub-agents, others fork them > extensions - SKILL.md in 9 of 11, MCP in 8 it ends with a 90-line harness covering all 7 the cleanest starting point for your own agent the pages worth your time: > p.4 - the 7-part map > p.12 - 3 loop types, Claude Code and Codex as examples > p.53 - why nobody uses frameworks or embeddings > p.67 - 18 design rules > p.71 - the 90-line harness to copy

this might be the most useful AI agent paper of 2026, and it's certified f*cking gold Wavestone AI Lab took apart Claude Code, Codex and 9 more agents every good one runs on the same 7-part harness agent = model + harness all 11 use the same 7 parts. only the size changes the 7 parts: > loop - from a plain while loop to fully replayable runs > LLM layer - one prompt template up to 29 provider profiles > tools - bash only up to 43 typed tools on demand > memory - full history up to notes kept across sessions > safety - a step limit up to rules + reviewer model + sandbox > orchestration - Aider skips sub-agents, others fork them > extensions - SKILL.md in 9 of 11, MCP in 8 it ends with a 90-line harness covering all 7 the cleanest starting point for your own agent the pages worth your time: > p.4 - the 7-part map > p.12 - 3 loop types, Claude Code and Codex as examples > p.53 - why nobody uses frameworks or embeddings > p.67 - 18 design rules > p.71 - the 90-line harness to copy

Annatar.md

50,060 views • 4 days ago

Here we see GPT-5.1 struggle to read a 500 lines markdown file for almost 2 minutes. Same model in a different harness does it in a few seconds (takes 2 tool calls for whatever reason..). Is all the Codex-love an elaborate joke I'm not in on?

Here we see GPT-5.1 struggle to read a 500 lines markdown file for almost 2 minutes. Same model in a different harness does it in a few seconds (takes 2 tool calls for whatever reason..). Is all the Codex-love an elaborate joke I'm not in on?

Mario Zechner

210,131 views • 10 months ago

Anthropic Technical Staff Jess Yan says the harness and the model cannot be pulled apart without giving up performance, though she flags her own bias. "I think I am quite biased but I also think that it is impossible to get the maximum possible performance without tying together the harness and the model." "Now the components of the harness and maybe the thickness of the harness will change over time as models get more and more capable." "However, when we test our models and when we are assessing their performance we always have to test it in conjunction with a harness." "And are we going to test it with all the different harnesses of the world? We're going to select the harnesses that we have built." "And so there is an aspect of the necessity of building models is that you have to be testing them with harnesses, and that sort of keeps them paired together."

Anthropic Technical Staff Jess Yan says the harness and the model cannot be pulled apart without giving up performance, though she flags her own bias. "I think I am quite biased but I also think that it is impossible to get the maximum possible performance without tying together the harness and the model." "Now the components of the harness and maybe the thickness of the harness will change over time as models get more and more capable." "However, when we test our models and when we are assessing their performance we always have to test it in conjunction with a harness." "And are we going to test it with all the different harnesses of the world? We're going to select the harnesses that we have built." "And so there is an aspect of the necessity of building models is that you have to be testing them with harnesses, and that sort of keeps them paired together."

Fireside Alpha

584,074 views • 2 months ago

Introducing Agent Mode: Agentic AI is now measured in the Arena. Agent Mode can do deep research, create reports, generate images, build websites, debug code, and more. It completes more complex tasks by using tools like web search, bash in a sandbox environment, image generation, file writing, and asking follow-up questions. Frontier models are waiting for you in Agent Mode to take on real-world tasks. GPT-5.5, Claude Opus 4.7, Gemini 3.1 Pro, and top open models. Test them yourself.

Introducing Agent Mode: Agentic AI is now measured in the Arena. Agent Mode can do deep research, create reports, generate images, build websites, debug code, and more. It completes more complex tasks by using tools like web search, bash in a sandbox environment, image generation, file writing, and asking follow-up questions. Frontier models are waiting for you in Agent Mode to take on real-world tasks. GPT-5.5, Claude Opus 4.7, Gemini 3.1 Pro, and top open models. Test them yourself.

Arena.ai

549,828 views • 4 months ago

Claude Code is the OpenClaw alternative you already had. I've been telling people this in person for three months. Now it's a piece on Every 📧. When OpenClaw was at peak hype a few months ago, everyone was going crazy over it — for good reasons. But almost nobody made the obvious comparison: Claude Code already does the thing you love about OpenClaw. In a lot of cases, better. Here's why people missed it. You have to understand the difference between a model and a harness. OpenClaw, at the end of the day, is just a harness on top of AI models. And Claude Code is too — it just got marketed as a coding tool, an alternative to Cursor. Until Cowork launched, almost nobody looked at it as a tool for non-programmers. Dan Shipper 📧 was one of the earliest to make that claim, and he was damn right. Let Claude Code run from the home folder of your computer with free access, and you get exactly what made OpenClaw exciting. Instead of just arguing it, here's a video walkthrough of a couple of my AI employees i built on the Claude Code harness. And here's the full article:

Claude Code is the OpenClaw alternative you already had. I've been telling people this in person for three months. Now it's a piece on Every 📧. When OpenClaw was at peak hype a few months ago, everyone was going crazy over it — for good reasons. But almost nobody made the obvious comparison: Claude Code already does the thing you love about OpenClaw. In a lot of cases, better. Here's why people missed it. You have to understand the difference between a model and a harness. OpenClaw, at the end of the day, is just a harness on top of AI models. And Claude Code is too — it just got marketed as a coding tool, an alternative to Cursor. Until Cowork launched, almost nobody looked at it as a tool for non-programmers. Dan Shipper 📧 was one of the earliest to make that claim, and he was damn right. Let Claude Code run from the home folder of your computer with free access, and you get exactly what made OpenClaw exciting. Instead of just arguing it, here's a video walkthrough of a couple of my AI employees i built on the Claude Code harness. And here's the full article:

Nityesh

22,895 views • 3 months ago

I made Claude compete against itself. The smartest model does not automatically make the best agent. To prove it, I took the same Opus 4.6 model, initial prompt, and empty repo and but it inside two different coding harnesses: Claude Code vs. Factory Droid. The task: clone Excalidraw from scratch, inspect the original in a browser, implement its key interactions, and verify the result. Factory Droid: 8 minutes, 20 tool calls, $1.60 Claude Code: 19 minutes, 40 tool calls, $1.87 The only difference was the harness. Get Free Factory credits here:

I made Claude compete against itself. The smartest model does not automatically make the best agent. To prove it, I took the same Opus 4.6 model, initial prompt, and empty repo and but it inside two different coding harnesses: Claude Code vs. Factory Droid. The task: clone Excalidraw from scratch, inspect the original in a browser, implement its key interactions, and verify the result. Factory Droid: 8 minutes, 20 tool calls, $1.60 Claude Code: 19 minutes, 40 tool calls, $1.87 The only difference was the harness. Get Free Factory credits here:

Sarah Chieng

104,774 views • 1 month ago

1/ First, how do you benchmark an AI model? You give different models the same test cases, and score them. You can do the same with voice models using Word Error Rate (WER). You generate the audio, transcribe it with a speech recognition (ASR) model, and compare that transcript to the original text. And this tells you whether the model said the right words. But the same sentence can be said correctly, while still being wrong in context.

1/ First, how do you benchmark an AI model? You give different models the same test cases, and score them. You can do the same with voice models using Word Error Rate (WER). You generate the audio, transcribe it with a speech recognition (ASR) model, and compare that transcript to the original text. And this tells you whether the model said the right words. But the same sentence can be said correctly, while still being wrong in context.

Cartesia

10,184 views • 23 days ago