OpenMLE: Open-Source AI4AI System Hits 71% on MLE-Bench Lite Using a Single RTX 4090
TianbaoX · x · 2026-07-31
FrontisAI launched OpenRSI, an open research series dedicated to recursive self-improvement (RSI), debuting with the full-stack AI4AI system OpenMLE. It uses evolutionary agents to improve ML solutions via executable feedback.
The system features three components: OpenMLE-Gym (5,758 executable tasks), OpenMLE-ERL (execution-grounded SFT + RL), and OpenMLE-Evo (experience-guided long-horizon search).
Performance: The trained Frontis-MA1-35B model paired with OpenMLE-Evo-Max achieves 71.21% Medal Average on MLE-Bench Lite, beating GPT-5.5 + Codex (68.18%) and closely trailing the top-tier Kimi K3 + Claude Code combo (72.73%). It is highly cost-effective, running on a single RTX 4090 (capped at 12GB VRAM) with a 12-hour per-task budget. The components also successfully transfer to 10 held-out NatureBench Lite tasks.
Related event: FrontisAI Open-Sources OpenMLE System and 35B Model(2 posts)→
More from coding & agent
- Burning a $100 Weekly Limit in One Prompt: The High Cost of Codex Testing — burny_tech · 2026-07-31
- Testing Inkling Small: A Vision-Equipped Model That Can Build Flappy Bird — LiTianleli · 2026-07-31
- Developer Uses Codex Ultra Mode to Bulk-Improve Agent Product, Adds Video Calls — jasonkneen · 2026-07-31
- Open Source Claude Code Skill: Generate Nothing's Minimalist UI Instantly — tom_doerr · 2026-07-31
- Don't Let LLMs Do Everything: When to Use Regex and Traditional ML — Ok_Philosophy_4031 · 2026-07-31
- Forge Orchestrator: Open-Source Tool to Coordinate Claude Code, Codex, and Gemini CLI — tom_doerr · 2026-07-31