Loading…
Loading…
Text generation · served with vLLM · ↓ 79.9K downloads a month · View on Hugging Face
llmat/Qwen3-4B-Instruct-2507-NVFP4 needs about 7.6 GB of GPU memory for conversations of up to 8,192 tokens. It runs on one RTX 3060 from $0.09/h right now; best value: RTX 5070 Ti at $0.22/h.
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 | Deploy |
|---|---|---|---|---|---|
| OriginalQwen/Qwen3-4B-Instruct-2507 | vLLM | 8.0 GB | 12.5 GB | $0.10/h RTX A4000 | Deploy |
| FP8Qwen/Qwen3-4B-Instruct-2507-FP8 | vLLM | 5.2 GB | 9.3 GB | $0.08/h RTX 3060 | Deploy |
| unsloth bnb 4bit (unsloth)unsloth/Qwen3-4B-Instruct-2507-unsloth-bnb-4bit | vLLM | 3.5 GB | 7.5 GB | $0.08/h RTX 3060 | Deploy |
| NVFP4 (llmat)llmat/Qwen3-4B-Instruct-2507-NVFP4 | vLLM | 3.6 GB | 7.6 GB | $0.08/h RTX 3060 | |
| Qwen3 Sex (ystemsrx)ystemsrx/Qwen3-Sex | vLLM | 8.0 GB | 12.5 GB |
$0.22/h from your balance while it runs, billed by the hour; stop it from its page. GridPup rents the machine, which downloads the model from Hugging Face (usually ready in 10 to 15 minutes), and gives you an https address, keys and a Playground.
| $0.10/h RTX A4000 |
| Deploy |
| bnb 4bit (unsloth)unsloth/Qwen3-4B-Instruct-2507-bnb-4bit | vLLM | 2.6 GB | 6.5 GB | $0.08/h RTX 3060 | Deploy |
| hpc assistant qwen3 4b (VaibhavKKM)VaibhavKKM/hpc-assistant-qwen3-4b | vLLM | 8.0 GB | 12.5 GB | $0.10/h RTX A4000 | Deploy |
| AWQ (warshanks)warshanks/Qwen3-4B-Instruct-2507-AWQ | vLLM | 3.4 GB | 7.4 GB | $0.08/h RTX 3060 | Deploy |
| onnx (ModaLeap)ModaLeap/qwen3-4b-instruct-2507-onnx | ONNX Runtime GenAI | 2.3 GB | 5.4 GB | $0.07/h GTX 1660 |
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 of 90% or more that download at 200 Mbit/s or more are offered: the machine fetches the model itself.