Olmo Hybrid 7B Think

Olmo Hybrid 7B Think needs an estimated 5.7 GB of VRAM at Q4, 9.6 GB at Q8 and 17.6 GB at FP16. All figures are estimates based on the 7.4B parameter count, weights plus about 15% and 0.5 GB for a short context.

Yes. A 24 GB GPU fits at FP16 (estimated 17.6 GB), a 16 GB or 12 GB GPU fits at Q8 (estimated 9.6 GB), and an 8 GB GPU fits at Q4 (estimated 5.7 GB). All figures are estimates.

Yes, at Q4 the model needs an estimated 5.7 GB. 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 Hybrid 7B Think has open weights, so you can self-host it for full control of your data.

Run Olmo Hybrid 7B Think locally

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

Closest alternatives, matched on model kind and release date with no comparable price: Qwen3.5 0.8B and Qwen3.5 2B by Alibaba, and LFM2 24B A2B by Liquid AI.

Qwen3.5 0.8BOlmo Hybrid 7B Think vs Qwen3.5 0.8BQwen3.5 2BOlmo Hybrid 7B Think vs Qwen3.5 2BLFM2 24B A2BOlmo Hybrid 7B Think vs LFM2 24B A2B

Olmo Hybrid 7B Think is open weights, not fully open source. The license is Apache 2.0.

Yes. Olmo Hybrid 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-Hybrid-Think-SFT-7B.

Run Olmo Hybrid 7B Think locally

Olmo Hybrid 7B Think was made by Ai2 and released on March 5, 2026.

More from Ai2

Yes. Olmo Hybrid 7B Think has open weights.