Thinking Machines Releases Inkling-Small: A 276B MoE Open Weight Model
LiTianleli · x · 2026-07-31
Thinking Machines Lab, founded by Mira Murati, has released Inkling-Small, an open-weight AI model.
- Architecture & Scale: A Mixture-of-Experts (MoE) model with 276B total parameters, activating only 12B parameters per token.
- Benchmarks: Outperforms the much larger 41B-active Inkling on Terminal-Bench 2.1, HLE, and SWE-Bench Verified.
- Technical Highlights: Using on-policy distillation and two weeks of coding RL, the smaller model outperforms its teacher on several benchmarks. It supports native audio and vision, features a 1M-token context window, and offers variable thinking effort.
Related event: Inkling-Small: 276B params, 12B active, open-source tops ARC(48 posts)→
More from Models
- Why are sub-15B open models so rare now? Labs seem to have settled on 24-27B minimum — Hot_Example_4456 · 2026-10-16
- User spots model openly generating "ideal responses" to train other models — liminal_bardo · 2026-10-16
- Reflection's 501B MoE Beam hits 463 tok/s on 8x GB200 with NVFP4 at day 0 — zhyncs42 · 2026-10-16
- Gemma 4 31B QAT on one 24GB card: 46 tok/s plain, ~80 with draft model on RTX 4090 — Subject_Poet_3861 · 2026-10-16
- JevBench v1.7.46 leaderboard: new d1-ultra tops 32 API systems at 76.1 composite — airesearch12 · 2026-10-16
- Team of 5 mulls dropping ChatGPT Pro for Business as OpenAI quota parity stays unclear — bettercallfool · 2026-10-16