Qwen3-Coder-Next-FP8
by Qwen
79.7B params · text-generation · 111 likes · 410.6k downloads
Qwen3-Coder-Next-FP8 is a 79.7B parameter model. At Q4 quantization it requires 40GB of VRAM. It runs comfortably on A100 PCIe 80 GB (30 tok/s), H100 SXM5 80 GB (60 tok/s), H200 SXM 141 GB (88 tok/s).
Inference providers
| Provider | $/1M in | $/1M out | Throughput |
|---|---|---|---|
| Together AI | 126 tok/s |
GPU compatibility
| GPU | VRAM | Q4 Decode | Verdict |
|---|---|---|---|
| M4 Max 128GB | 128GB | 6 tok/s | tight |
| M4 Pro 48GB | 48GB | 3 tok/s | tight |
| A100 PCIe 80 GB | 80GB | 30 tok/s | comfortable |
| H100 SXM5 80 GB | 80GB | 60 tok/s | comfortable |
| M4 Max 64GB | 64GB | 6 tok/s | tight |
| H200 SXM 141 GB | 141GB | 88 tok/s | comfortable |
| M2 Ultra 192GB | 192GB | 8 tok/s | tight |