Thinking Machines Lab Releases Inkling-Small: A 276B Total, 12B Active Open Weights Multimodal MoE Model
Thinking Machines Lab has released Inkling-Small, an open weights Mixture-of-Experts model with 276B total parameters and 12B active. That is about a quarter the size of Inkling, which carries 975B total and 41B active parameters. The model was trained on NVIDIA GB300 NVL72 systems. It reasons natively over text, images and audio. The context window reaches 1M tokens, and thinking effort is adjustable. Weights ship under Apache 2.0 on Hugging Face. Is it deployable Yes, and the quantized
Get smart on it
Thinking Machines Lab has released Inkling-Small, an open weights model with 276 billion total parameters but only 12 billion active parameters at any given time. This matters because the model can run on a single high-end GPU instance, making it accessible to startups and enterprises that cannot afford frontier-scale hardware, while still handling text, images, and audio inputs simultaneously. The model uses a sparse mixture-of-experts architecture where each token routes through only a small subset of its available experts, reducing computational cost compared to using all parameters. On several reasoning and coding benchmarks, the smaller model outperformed its larger counterpart despite being a quarter of its size, though it showed weaker performance on factual recall tasks.
Related news
Pokee AI Releases Pokee-Isaac 28B: A 10M-Token Context Agentic Model Built to Run Inside the Customer Boundary
Long-horizon agents accumulate context faster than they resolve tasks. Every tool output, observation, and intermediate reasoning step stays in the window, and the two capabilities that matter — holding that context and staying coherent across it — have so far been available almost exclusively from cloud endpoints. That excludes regulated industries, public-sector institutions, and on-device applications, where the data is not permitted to leave the boundary at all. Pokee AI released Pokee-Isaa
Mistral AI Releases Shieldstral 1.0 3B: An Open-Weights Policy-Adaptive Multimodal Safety Classifier Matching Models 7× Its Size
Mistral AI has released Shieldstral 1.0 3B, an open-weights, policy-adaptive multimodal safety classifier that treats content moderation as a single yes/no question rather than a fixed taxonomy of harm categories. Most guardrail models bake their category list into the weights, so re-targeting one to a new deployment context means retraining — and the same content can be acceptable on a cybersecurity research tool while being harmful on a mental-health platform. Shieldstral inverts that: operat
