25B-token GPT-5.5 dataset with 12,121 tool calls released on Hugging Face
-MaskNinja- · reddit · 2026-11-09
A Reddit user released a GPT-5.5 dataset built from 25 billion tokens of personal usage, including 12,121 tool-call trajectories. The data was sanitized with help from another user and published on Hugging Face as israelllaguan/codex-tool-use-trajectories — a directly downloadable resource for anyone studying agent tool use or fine-tuning on real trajectories.
More from Research
- New theory shows prover-conjecturer self-play can grow proved theorems exponentially — StanfordAILab · 2026-11-09
- AbridgeKV: latent KV cache compression matches full accuracy at 1/4 cache, 3.2× faster — DimitrisPapail · 2026-11-09
- Stanford's Hinted Scaling Laws cut LLM capability forecast error from 21% to 4.3% — burny_tech · 2026-11-09
- Stanford's Evo AI designs 16 working bacteria-killing viruses, sparking access debate — coursiv_ · 2026-11-09
- FOCS 2026 opens in NYC, jokingly called the last conference with human-proven theorems — fortnow · 2026-11-09
- NVIDIA's SIGGRAPH 2026 paper DAT delivers penetration-free contact for cloth, soft bodies and robots — zhengyiluo · 2026-11-09