Loading…
Loading…
Models / Qwen3.8-Flash-Next-W4A16-AutoRound
1 model (~124B) · ↓ 10.9K downloads a month · from $0.90/h on 2x RTX A6000 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 |
|---|---|---|---|---|
| Qwen3.8 Flash Next 125B A5B INT4 AutoRound (azampatti)azampatti/Qwen3.8-Flash-Next-125B-A5B-INT4-AutoRound | vLLM | 75 GB | 86.7 GB | $0.90/h 2x RTX A6000 |
| Qwen3.8 Flash Next W4A16 FP8PLE (albucino)albucino/Qwen3.8-Flash-Next-W4A16-FP8PLE | vLLM | 125 GB | 141.8 GB | $1.44/h 8x RTX 3090 |
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").