PostTrainBench one year on: AI agents now run complex post-training pipelines
maksym_andr · x · 2026-11-17
One year after releasing PostTrainBench, which measures how well agents like Claude Code can post-train base LLMs, maksymandr reports progress outpaced expectations: frontier models went from barely launching simple SFT runs to reliably implementing complex (if not yet novel) post-training pipelines. The leaderboard shows CLI agents powered by various models scoring from 18.1% (Gemini 3.1 Pro) up to 44.6% (Fable 5.2 Max) vs an 8.0% base-model baseline, with official instruct models at 48.4% — a key indicator for AI R&D automation.
More from coding & agent
- Ex-fintech insider: agentic payments UX is an open market waiting to be owned — SimonBalmain · 2026-11-17
- He automated his writing with a Claude Skill — then used it to write about itself — danshipper · 2026-11-17
- Tencent Open-Sources LoopForge, a Self-Hosted Self-Evolving AI Software Delivery Platform — aigclink · 2026-11-17
- Running a language model on a 1979 CPU: Tiny 2.1 ported to the Amiga 500 — azhadron · 2026-11-17
- Berkeley and Microsoft paper cuts coding-agent token costs 18.4-30.9% with a black-box external controller — rohanpaul_ai · 2026-11-17
- Your voice agent got flagged 'Spam Likely'? Swapping the DID is the mistake — Packet-Exchange-Care · 2026-11-17