Surge RL-Tuned Nemotron 3.5 Super Beats Models Up to 20x Its Size
echen · x · 2026-11-06
Surge AI researcher echen argues people are underestimating open models' headroom: the team trained NVIDIA Nemotron 3.5 Super (12B active) with Surge RL data via Thinking Machines' Tinker, beating much larger closed models across 12 benchmarks including GPT-5.6 Terra on coding.
The accompanying chart shows Nemotron + Surge RL reaching top ranks: GDP.pdf +18.6pp, HANDBOOK.md +12.8pp, AutomationBench +17.6pts (31.9→49.5), Chartography +11.2pp, AdvancedIF +7.7pp, Terminal-Bench 2.1 +6.4pp (63.9%→70.3%, past Claude Opus 4.8 and GLM 5.3), DeepSWE +4.7pp.
Takeaway: high-quality RL data makes frontier performance accessible to everyone.
Related event: Surge AI RL-Trains Nemotron 3.5 Super on Tinker, Nearing GPT-5.6(6 posts)→
More from coding & agent
- Google open-sources Agile Deliberation, an AI framework that turns fuzzy classification rules into explicit ones — aigclink · 2026-11-06
- AI uncovers 10-year-old bug where any text over 256 chars works as a valid root password — andersonbcdefg · 2026-11-06
- vLLM open-sources System Duo: a fast/slow semantic router that only calls deep reasoning when needed — vllm_project · 2026-11-06
- "Throw inference at problems until they go away" — a workflow take on scaling compute — rmcwhorter99 · 2026-11-06
- Claude Code bug leaks user email via User-Agent header, victim reports dozens of scam emails daily — niloofar_mire · 2026-11-06
- Nous Research ships Hermes Agent v0.23.0 with desktop-in-browser and unified gateway — NousResearch · 2026-11-06