Loading…
Loading…
Models / Step3-VL
1 model (10B) · ↓ 14.8K downloads a month · from $0.31/h on 2x RTX 4060 Ti 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 |
|---|---|---|---|---|
| Instructstepfun-ai/Step3-VL-10B | vLLM | 20 GB | 26.2 GB | $0.31/h 2x RTX 4060 Ti |
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").