Largest open experiment yet on autonomous agents iterating in a research environment
eliebakouch · x · 2026-09-09
eliebakouch announces what he claims is the largest open experiment on autonomous agents iterating on a research environment, scaling runtime, compute, and diversity of models and harnesses.
For comparison, similar tasks in system cards — Anthropic's "optimizing an LLM training on CPU" and OpenAI's GPT-5.6 on nanogpt track 1 — ran for less than a day. The project shares extensive details including traces and scratchpads, with more results coming.
More from coding & agent
- Researchers formally deny AI agents peeked at prior work before solving the problem — zedlander · 2026-09-09
- Agent swarm burns 300B tokens in 88 hours to crack Navier-Stokes, dev claims — bingxu_ · 2026-09-09
- LangChain details two context modes for subagents: isolated vs forked — LangChain · 2026-09-09
- Cheap models via OpenRouter fall apart in agentic harnesses: GLM and DeepSeek can't match Claude — scottyLogJobs · 2026-09-09
- Multi-agent coding's hardest problem: deciding who is allowed to change what — apghere · 2026-09-09
- One creator's 3-stage AI video pipeline: GPT-6 Astra plans, Seedance 2.5 shoots, CapCut's Edit Pilot edits — azed_ai · 2026-09-09