MIT Releases FinanceGym, the Largest Finance-Agent Dataset with 59K+ Samples
MIT_CSAIL · x · 2026-08-25
MIT CSAIL researchers have released "FinanceGym," the largest finance-agent dataset to date, featuring over 59,000 samples across 37 roles and 40 apps/sites.
The data can be used for Reinforcement Finetuning or Supervised Finetuning of agentic models. It contains human-demonstrated trajectories with thinking traces, designed to teach agents financial work rather than just financial knowledge. Tests show that even Kimi K3 and Qwen 3.8 Max failed its test set.
More from coding & agent
- Engineer: my job went from silently thinking at a computer to constantly talking to it — KevinNaughtonJr · 2026-10-16
- Vercel Goes Full Backend: Python GA, Containers, and Daemons — cramforce · 2026-10-16
- WorldMemArena Accepted at NeurIPS: Memory Storage ≠ Memory Use — xwang_lk · 2026-10-16
- Vercel Ship Demos Container Deploys for Go, Ruby, Java and More — edwin · 2026-10-16
- TRL adds day-0 support for fine-tuning Muse Spark Open's 816B MoE with LoRA on 64 H100s — QGallouedec · 2026-10-16
- Databricks engineers land impactful PRs on day 2 with AI dev tools, vs week 3 before — Yuchenj_UW · 2026-10-16