Thinking Machines releases Inkling, a 975B open-weights multimodal model with 1M context
paraschopra · x · 2026-07-28
Thinking Machines has announced Inkling, an open-weights foundation model trained from scratch and released with full weights.
- Architecture: a Mixture-of-Experts transformer with 975B total parameters and 41B active.
- Context window: up to 1M tokens.
- Training data: 45T tokens across text, images, audio, and video.
- Product strategy: Inkling is positioned as a customizable base model rather than the strongest model overall, and is available for fine-tuning on Tinker.
- Family roadmap: the company also previewed Inkling-Small, a 12B-active variant aimed at lower cost and latency.
The release emphasizes multimodal native reasoning, efficient/controllable thinking effort, and a broader push to make customization easier through the Inkling Playground in Tinker.
More from Models
- Claude meme turns evals into poetry, joking about 0.03 deceptive alignment — maxsloef · 2026-07-28
- Gemini Flash 3.6 is claimed to match Sol 5.6 quality at 70% lower cost — bindureddy · 2026-07-28
- Anthropic’s Opus 4.8 beats version 5 for non-coding work, despite weaker benchmarks — burkov · 2026-07-28
- Users Say Opus 5 Looks Better After Repeated Bug-Fix and Feature Requests — Rasmic · 2026-07-28
- A user says GPT-5.4, Opus 4.6, and Kimi k3 already cover most needs — haider1 · 2026-07-28
- LLaDA2.2 brings diffusion language models into long-horizon agent tasks — 量子位 · 2026-07-28