Olmo 3 models
8 Olmo 3 models on Hugging Face, from 7.3B to 32.2B parameters. At the precision each one is published in, the smallest needs about 16.3 GB of VRAM (Olmo-3-1025-7B, cheapest live fit: RTX A5000) and the largest about 72.0 GB (Olmo-3-1125-32B, cheapest live fit: A100). The cheapest way to run Olmo-3-1025-7B is $0.176/hr.
Olmo 3 models
| Model | Parameters | Published as | VRAM needed | Live GPU fit | Est. $/hr |
|---|---|---|---|---|---|
| Olmo-3-1025-7B | 7.3B | BF16 | 16.3 GB | RTX A5000 | $0.176/hr |
| Olmo-3-7B-Instruct | 7.3B | BF16 | 16.3 GB | RTX A5000 | $0.176/hr |
| Olmo-3-7B-Instruct-DPO | 7.3B | BF16 | 16.3 GB | RTX A5000 | $0.176/hr |
| Olmo-3-7B-Instruct-SFT | 7.3B | BF16 | 16.3 GB | RTX A5000 | $0.176/hr |
| Olmo-3-7B-Think | 7.3B | BF16 | 16.3 GB | RTX A5000 | $0.176/hr |
| Olmo-3-7B-Think-SFT | 7.3B | BF16 | 16.3 GB | RTX A5000 | $0.176/hr |
| Olmo-3-1125-32B | 32.2B | BF16 | 72.0 GB | A100 | $1.21/hr |
| Olmo-3.1-32B-Instruct | 32.2B | BF16 | 72.0 GB | A100 | $1.21/hr |
VRAM is for the precision the model is published in, with the same overhead and an 8,192-token context assumed on every page; see the methodology. The fit shown is the lowest-priced single GPU type that holds the model at that precision, or the lowest-priced multi-GPU set (up to 8) when none does.
