Thinking Machines Launches Inkling-Small: A 276B MoE Model with Variable Thinking Effort
clarejtbirch · x · 2026-07-31
Thinking Machines Lab has released Inkling-Small, an efficient open-weights Mixture-of-Experts (MoE) model. It features 276B total parameters with 12B active, trained on NVIDIA GB300 NVL72 systems.
Inkling-Small delivers performance comparable to significantly larger models but at a fraction of the compute cost. It supports native reasoning over audio and images, offers variable thinking effort, and boasts a context window of up to 1M tokens. Across benchmarks for agentic tool use, reasoning, and instruction following, the model demonstrates exceptional cost-efficiency, allowing users to easily balance cost and performance by adjusting reasoning effort.
Related event: Inkling-Small: 276B params, 12B active, open-source tops ARC(48 posts)→
More from Models
- Reflection's Beam lands on Hugging Face: 501B MoE, 1M context, Apache 2.0 open weights — tomaarsen · 2026-10-16
- Reflection launches Beam 1.0, its first open-weight model built for coding, reasoning and agentic tasks — tomaarsen · 2026-10-16
- Reflection's 501B MoE Beam hits 463 tok/s on 8x GB200 with NVFP4 at day 0 — zhyncs42 · 2026-10-16
- JevBench v1.7.46 leaderboard: new d1-ultra tops 32 API systems at 76.1 composite — airesearch12 · 2026-10-16
- LMArena launches Agent Arena and Code Arena leaderboards for model comparison — arena · 2026-10-16
- Third-party tests find NVIDIA Nemotron-3.5 Bolt underperforms nano on pandas/numpy and decision-making tasks — norenEnmotalen · 2026-10-16