Loading…
Loading…
Models / Qwen3.5
4 models (4B to 122B) · ↓ 53K 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 |
|---|---|---|---|---|
| Originalz-lab/Qwen3.5-4B-DFlash | vLLM | 1.3 GB | 3.9 GB | $0.10/h RTX 3060 |
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|---|---|---|---|
| Originalz-lab/Qwen3.5-9B-DFlash | vLLM | 2.6 GB | 5.3 GB | $0.10/h RTX 3060 |
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|---|---|---|---|
| Originalz-lab/Qwen3.5-27B-DFlash | vLLM | 4.3 GB | 7.2 GB | $0.10/h RTX 3060 |
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|---|---|---|---|
| Originalz-lab/Qwen3.5-122B-A10B-DFlash | vLLM | 1.6 GB | 4.2 GB | $0.10/h RTX 3060 |
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").