Loading…
Loading…
Models / Mistral
3 models (7B to 7B) · ↓ 2.6M downloads a month · from $0.10/h on RTX 3060 right now. Each model's versions do the same job; quantized ones (AWQ, FP8, 4-bit) need less GPU memory and run on cheaper machines.
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|---|---|---|---|
| Originalmistralai/Mistral-7B-Instruct-v0.2 | vLLM | 14 GB | 19.5 GB | $0.18/h 2x RTX 3060 |
| AWQ (TheBloke)TheBloke/Mistral-7B-Instruct-v0.2-AWQ | vLLM | 4.2 GB | 8 GB | $0.10/h RTX 3060 |
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|---|---|---|---|
| Originalmistralai/Mistral-7B-v0.1 | vLLM | 14 GB | 19.5 GB | $0.18/h 2x RTX 3060 |
| Mistral 7B Instruct v0.1mistralai/Mistral-7B-Instruct-v0.1 | vLLM | 14 GB | 19.5 GB | $0.18/h 2x RTX 3060 |
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|---|---|---|---|
| bnb 4bit (unsloth)unsloth/mistral-7b-v0.3-bnb-4bit | vLLM | 4.1 GB | 8 GB | $0.10/h RTX 3060 |
| Mistral 7B Instruct v0.3 AWQ (solidrust)solidrust/Mistral-7B-Instruct-v0.3-AWQ | vLLM | 4.2 GB | 8 GB | $0.10/h RTX 3060 |
| mistral 7b instruct v0.3 bnb 4bit (unsloth)unsloth/mistral-7b-instruct-v0.3-bnb-4bit | vLLM | 4.1 GB | 8 GB | $0.10/h RTX 3060 |
| Mistral 7B Instruct v0.3 (Prince-1)Prince-1/Mistral-7B-Instruct-v0.3 | ONNX Runtime GenAI | 15 GB | 19.3 GB | $0.18/h Nvidia V100 |
| Mistral 7B Instruct v0.3 (onnx-community)onnx-community/Mistral-7B-Instruct-v0.3 | ONNX Runtime GenAI | 15 GB | 19.3 GB | $0.18/h Nvidia V100 |
Model details come from Hugging Face and are refreshed daily; prices follow the live offers. GPU memory is GridPup's estimate: the weights, room for an 8,192-token conversation (chat models) and the runtime's own memory. Large models are split across the GPUs of one machine ("4x RTX 4090").