Bolmo 7B

At Q4 Bolmo 7B needs an estimated 5.8 GB of VRAM, at Q8 an estimated 9.8 GB, and at FP16 an estimated 18.1 GB. These figures are estimates covering the weights plus about 15% and 0.5 GB for a short context.

Yes. A 24 GB GPU fits FP16 (estimated 18.1 GB), a 16 GB or 12 GB GPU fits Q8 (estimated 9.8 GB), and an 8 GB GPU fits Q4 (estimated 5.8 GB). All figures are estimates.

Yes. Assuming macOS lets the GPU use about 75% of unified memory, Q4 needs an estimated 5.8 GB, and the smallest common Mac memory size that fits that is 8 GB.

Yes. Bolmo 7B has open weights, so you can self-host it for full control of your data.

Run Bolmo 7B locally

No. Bolmo 7B only accepts text input.

MiMo-V2-Flash by Xiaomi, AutoGLM Phone 9B Multilingual by Z.ai and Devstral Small 2 by Mistral AI. They are matched on model kind and release date, with no comparable price.

MiMo-V2-FlashBolmo 7B vs MiMo-V2-FlashAutoGLM Phone 9B MultilingualBolmo 7B vs AutoGLM Phone 9B MultilingualDevstral Small 2Bolmo 7B vs Devstral Small 2

Open weights. Bolmo 7B is released under the Apache 2.0 license.

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

The Hugging Face repository allenai/Bolmo-7B hosts the Bolmo 7B weights.

Run Bolmo 7B locally

Ai2 made Bolmo 7B, and it was released on December 15, 2025.

More from Ai2

Yes. Bolmo 7B is open weights, so there is no charge for the model itself.