OLMo 2 models
5 OLMo 2 models from Ai2 on Hugging Face, from 1.4B to 32.2B parameters, published by Ai2 in F32 and BF16, with a 4K-token context. The smallest official model, OLMo-2-0425-1B-Instruct, needs about 3.3 GB of VRAM at its published precision; the cheapest live fit is RTX 3070 at $0.088/hr.
Part of the OLMo series · Previous generation: OLMo 1 · Next generation: Olmo 3
Pick a size
One row per official OLMo 2 size: the VRAM it needs at each precision, the cheapest GPU that holds it today, and the KV cache for a 32K-token context.
| Model | Parameters | Native VRAM | FP8 VRAM | INT4 VRAM | Live GPU fit (native) | Est. $/hr | KV cache at 32K |
|---|---|---|---|---|---|---|---|
| OLMo-2-0425-1B | 1.5B | 6.6 GB | 1.7 GB | 0.8 GB | RTX 3070 | $0.088/hr | 4.0 GB |
| OLMo-2-1124-7B-Instruct | 7.3B | 16.3 GB | 8.2 GB | 4.1 GB | RTX A5000 | $0.176/hr | 16.0 GB |
| OLMo-2-0325-32B-Instruct | 32.2B | 72.0 GB | 36.0 GB | 18.0 GB | A100 | $1.31/hr | 8.0 GB |
VRAM is weights times a flat 1.2 overhead; the KV cache is a separate per-model figure at 16-bit, shown where the architecture is published. See the methodology.
Official models (4)
| Model | Parameters | Published as | VRAM needed | Live GPU fit | Est. $/hr |
|---|---|---|---|---|---|
| OLMo-2-0425-1B | 1.5B | F32 | 6.6 GB | RTX 3070 | $0.088/hr |
| OLMo-2-0425-1B-Instruct | 1.5B | BF16 | 3.3 GB | RTX 3070 | $0.088/hr |
| OLMo-2-1124-7B-Instruct | 7.3B | BF16 | 16.3 GB | RTX A5000 | $0.176/hr |
| OLMo-2-0325-32B-Instruct | 32.2B | BF16 | 72.0 GB | A100 | $1.31/hr |
Fine-tunes and community models (1)
| Model | Parameters | Published as | VRAM needed | Live GPU fit | Est. $/hr |
|---|---|---|---|---|---|
| Recurrent-OLMo-2-0425-train-recurrence-16 | 1.4B | F32 | 6.1 GB | RTX 3070 | $0.088/hr |
VRAM is for the precision the model is published in: the weights times a flat 1.2 overhead, with the KV cache not included. 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.