Olmo 3 7B Think

Olmo 3 7B Think has 7.3B parameters. Estimated VRAM needs are about 5.6 GB at Q4, 9.4 GB at Q8 and 17.3 GB at FP16. These figures are estimates from the parameter count, covering weights plus about 15% and 0.5 GB for a short context.

Yes. On a 24 GB GPU it fits at FP16 (estimated 17.3 GB), on 16 GB and 12 GB GPUs it fits at Q8 (estimated 9.4 GB), and on an 8 GB GPU it fits at Q4 (estimated 5.6 GB).

Yes. At Q4 the estimated VRAM is 5.6 GB, and assuming macOS lets the GPU use about 75% of unified memory, the smallest common Mac memory size that fits Q4 is 8 GB (estimate).

Yes. Olmo 3 7B Think has open weights, which lets you self-host it for full control of your data.

Run Olmo 3 7B Think locally

No. Olmo 3 7B Think accepts text input only.

The closest alternatives, matched on model kind and release date with no comparable price, are ERNIE 4.5 VL 28B A3B Thinking by Baidu, DeepSeek Reasoner by DeepSeek and Trinity Mini by Arcee AI.

ERNIE 4.5 VL 28B A3B ThinkingOlmo 3 7B Think vs ERNIE 4.5 VL 28B A3B ThinkingDeepSeek ReasonerOlmo 3 7B Think vs DeepSeek ReasonerTrinity MiniOlmo 3 7B Think vs Trinity Mini

Olmo 3 7B Think is open weights. It is released under the Apache 2.0 license.

Yes. Olmo 3 7B Think uses the Apache 2.0 license, a permissive license that allows commercial use.

You can download the weights from the Hugging Face repository allenai/Olmo-3-7B-Think.

Run Olmo 3 7B Think locally

Olmo 3 7B Think was made by Ai2 and released on November 20, 2025.

More from Ai2

Yes. Olmo 3 7B Think has open weights.