Loading…
Loading…
Other · served with ONNX Runtime GenAI · ↓ 0 downloads a month · View on Hugging Face
The same model packaged differently: a quantized version needs less GPU memory and runs on a cheaper machine.
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|---|---|---|---|
| OriginalQwen/Qwen2.5-7B | vLLM | 15 GB | 19.7 GB | $0.18/h 2x RTX 3060 |
| InstructQwen/Qwen2.5-7B-Instruct | vLLM | 15 GB | 19.7 GB | $0.18/h 2x RTX 3060 |
| Instruct AWQQwen/Qwen2.5-7B-Instruct-AWQ | vLLM | 5.6 GB | 8.9 GB | $0.10/h RTX 3060 |
| Qwen2.5 Coder 7B InstructQwen/Qwen2.5-Coder-7B-Instruct | vLLM | 15 GB | 19.7 GB | $0.18/h 2x RTX 3060 |
| Qwen2.5 Coder 7B Instruct AWQQwen/Qwen2.5-Coder-7B-Instruct-AWQ | vLLM | 5.6 GB | 8.9 GB | $0.10/h RTX 3060 |
| Qwen2.5 Coder 7B Instruct GPTQ Int4Qwen/Qwen2.5-Coder-7B-Instruct-GPTQ-Int4 | vLLM | 5.6 GB | 8.9 GB | $0.10/h RTX 3060 |
| Instruct GPTQ Int4Qwen/Qwen2.5-7B-Instruct-GPTQ-Int4 | vLLM | 5.6 GB | 8.9 GB | $0.10/h RTX 3060 |
| Qwen2.5 Coder 7BQwen/Qwen2.5-Coder-7B | vLLM | 15 GB | 19.7 GB | $0.18/h 2x RTX 3060 |
| Qwen2.5 Math 7B InstructQwen/Qwen2.5-Math-7B-Instruct | vLLM | 15 GB | 19.4 GB | $0.18/h 2x RTX 3060 |
| Instruct 1MQwen/Qwen2.5-7B-Instruct-1M | vLLM | 15 GB | 19.7 GB | $0.18/h 2x RTX 3060 |
| Qwen2.5 Math 7BQwen/Qwen2.5-Math-7B | vLLM | 15 GB | 19.4 GB | $0.18/h 2x RTX 3060 |
| Qwen2.5 Coder 7B Instruct AWQ (Orion-zhen)Orion-zhen/Qwen2.5-Coder-7B-Instruct-AWQ | vLLM | 5.7 GB | 9.1 GB | $0.10/h RTX 3060 |
| Instruct bnb 4bit (unsloth)unsloth/Qwen2.5-7B-Instruct-bnb-4bit | vLLM | 5.5 GB | 8.9 GB | $0.10/h RTX 3060 |
| Qwen2.5 Coder 7B Instruct bnb 4bit (unsloth)unsloth/Qwen2.5-Coder-7B-Instruct-bnb-4bit | vLLM | 5.5 GB | 8.9 GB | $0.10/h RTX 3060 |
| Instruct unsloth bnb 4bit (unsloth)unsloth/Qwen2.5-7B-Instruct-unsloth-bnb-4bit | vLLM | 7.2 GB | 10.7 GB | $0.10/h RTX 3060 |
| Instruct FP8 dynamic (RedHatAI)RedHatAI/Qwen2.5-7B-Instruct-FP8-dynamic | vLLM | 8.7 GB | 12.4 GB | $0.16/h RTX 4060 Ti |
| Instruct heretic (saidutta69)saidutta69/Qwen2.5-7B-Instruct-heretic | vLLM | 15 GB | 19.7 GB | $0.18/h 2x RTX 3060 |
| Qwen2.5 Coder 7B bnb 4bit (unsloth)unsloth/Qwen2.5-Coder-7B-bnb-4bit | vLLM | 5.5 GB | 8.9 GB | $0.10/h RTX 3060 |
| Instruct ONNX WebGPU (danyalmalik)danyalmalik/Qwen2.5-7B-Instruct-ONNX-WebGPU | ONNX Runtime GenAI | 5.2 GB | 8 GB | $0.10/h RTX 3060 |
| Qwen2.5 Coder 7B Instruct ONNX WebGPU (danyalmalik)danyalmalik/Qwen2.5-Coder-7B-Instruct-ONNX-WebGPU | ONNX Runtime GenAI | 5.2 GB | 8 GB | $0.10/h RTX 3060 |
Deploying ONNX models from the catalog is coming soon. Today, download the files from Hugging Face and upload them in Inference.
Prices are per hour for the whole machine and its disk, right now. Speed is Vast's DLPerf score of the machine, relative to the cheapest. Best value: the most speed per dollar. Only machines with a Vast reliability above 95% that download at 400 Mbit/s or more are offered: the machine fetches the model itself.