OLMo 1B

OLMo 1B needs an estimated 1.3 GB of VRAM at Q4, 1.9 GB at Q8 and 3.2 GB at FP16. The figures are estimates from the 1.2B parameter count, weights plus about 15% and 0.5 GB for a short context.

Yes. OLMo 1B fits at FP16 on a 24 GB, 16 GB, 12 GB or 8 GB GPU, with an estimated 3.2 GB of VRAM.

Yes. At Q4 OLMo 1B needs an estimated 1.3 GB. Assuming macOS lets the GPU use about 75% of unified memory, the smallest common Mac memory size that fits is 8 GB.

Yes. OLMo 1B has open weights, so you can self-host it for full control of your data.

Run OLMo 1B locally

No. OLMo 1B supports text input only.

The closest alternatives, matched on model kind and release date with no comparable price, are BGE Reranker V2 M3 by BAAI, SOLAR 10.7B Instruct v1.0 by Upstage and Mixtral 8x7B Instruct v0.1 by Mistral AI.

BGE Reranker V2 M3OLMo 1B vs BGE Reranker V2 M3SOLAR 10.7B Instruct v1.0OLMo 1B vs SOLAR 10.7B Instruct v1.0Mixtral 8x7B Instruct v0.1OLMo 1B vs Mixtral 8x7B Instruct v0.1

OLMo 1B is open weights, released under the Apache 2.0 license.

Yes. OLMo 1B is licensed under Apache 2.0, a permissive license that allows commercial use.

You can download the OLMo 1B weights from the allenai/OLMo-1B repository on Hugging Face.

Run OLMo 1B locally

OLMo 1B was made by Ai2 and released on February 1, 2024.

More from Ai2

Yes. OLMo 1B has open weights.