Mistral 7B models
7 Mistral 7B models from Mistral AI on Hugging Face, from 7.2B to 7.2B parameters, published by Mistral AI in BF16, with a 32K-token context. The smallest official model, Mistral-7B-Instruct-v0.1, needs about 16.2 GB of VRAM at its published precision; the cheapest live fit is RTX A5000 at $0.176/hr.
Part of the Mistral series · Next generation: Mixtral
Pick a size
One row per official Mistral 7B 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 |
|---|---|---|---|---|---|---|---|
| Mistral-7B-Instruct-v0.2 | 7.2B | 16.2 GB | 8.1 GB | 4.0 GB | RTX A5000 | $0.176/hr | 4.0 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 (3)
| Model | Parameters | Published as | VRAM needed | Live GPU fit | Est. $/hr |
|---|---|---|---|---|---|
| Mistral-7B-Instruct-v0.1 | 7.2B | BF16 | 16.2 GB | RTX A5000 | $0.176/hr |
| Mistral-7B-Instruct-v0.2 | 7.2B | BF16 | 16.2 GB | RTX A5000 | $0.176/hr |
| Mistral-7B-v0.1 | 7.2B | BF16 | 16.2 GB | RTX A5000 | $0.176/hr |
Fine-tunes and community models (4)
| Model | Parameters | Published as | VRAM needed | Live GPU fit | Est. $/hr |
|---|---|---|---|---|---|
| GritLM-7B | 7.2B | BF16 | 16.2 GB | RTX A5000 | $0.176/hr |
| zephyr-7b-beta | 7.2B | BF16 | 16.2 GB | RTX A5000 | $0.176/hr |
| neural-chat-7b-v3-3 | 7.2B | F16 | 16.2 GB | RTX A5000 | $0.176/hr |
| openchat-3.5-0106 | 7.2B | BF16 | 16.2 GB | RTX A5000 | $0.176/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.
More from Mistral AI
- Voxtral (1 models)
- Mistral Small (3 models)
- Mixtral (1 models)
- mistralai (1 models)