Olmo 3.1 32B Instruct

Olmo 3.1 32B Instruct is estimated to need about 23 GB of VRAM at Q4, 39.9 GB at Q8 and 74.6 GB at FP16. These figures are estimates based on the 32.2B parameter count, covering weights plus about 15% and 0.5 GB for a short context.

Yes on a 24 GB GPU, it fits at Q4 with an estimated 23 GB. No on 16 GB, 12 GB or 8 GB GPUs, which do not fit at Q4, Q8 or FP16 (estimate).

Yes, with enough unified memory. At Q4 the estimated need is 23 GB, and assuming macOS lets the GPU use about 75% of unified memory, the smallest common Mac memory size that fits is 32 GB (estimate).

Yes. Olmo 3.1 32B Instruct has open weights, so you can self-host it for full control of your data.

Run Olmo 3.1 32B Instruct locally

No. Olmo 3.1 32B Instruct accepts text input only, not images, audio, video or PDFs.

Olmo 3.1 32B Instruct holds rank 251 on Arena Text (Arena's own rank) with a score of 1329.

Olmo 3.1 32B Instruct benchmarks

The closest alternatives, matched on model kind and release date with no comparable price, are AutoGLM Phone 9B Multilingual by Z.ai, Devstral Small 2 by Mistral AI and GLM-4.6V-Flash by Z.ai.

AutoGLM Phone 9B MultilingualOlmo 3.1 32B Instruct vs AutoGLM Phone 9B MultilingualDevstral Small 2Olmo 3.1 32B Instruct vs Devstral Small 2GLM-4.6V-FlashOlmo 3.1 32B Instruct vs GLM-4.6V-Flash

Olmo 3.1 32B Instruct has open weights rather than full open source. Its license is Apache 2.0.

Yes. Olmo 3.1 32B Instruct uses the Apache 2.0 license, a permissive license that allows commercial use.

The weights are available on the Hugging Face repository allenai/Olmo-3.1-32B-Instruct.

Run Olmo 3.1 32B Instruct locally

Olmo 3.1 32B Instruct was made by Ai2 and released on December 12, 2025.

More from Ai2

Yes. Olmo 3.1 32B Instruct has open weights.