Qwen3-Coder-480B-A35B — 480.2B Parameter Mixture of Experts LLM
Model Specifications
- Parameters
- 480.2B (35B active)
- Architecture
- Mixture of Experts
- Context Length
- 256K tokens
- Capabilities
- chat, tool_use, coding
- Release Date
- 2025-11-01
- Provider
- Alibaba
- Family
- qwen
VRAM Requirements
| Quantization | BPW | VRAM | Quality |
|---|---|---|---|
| IQ2_XXS | 2.38 | 143.3 GB | 65% |
| IQ2_M | 2.93 | 176.4 GB | 75% |
| Q2_K | 3.16 | 190.2 GB | 78% |
| IQ3_XXS | 3.25 | 195.6 GB | 82% |
| IQ3_XS | 3.5 | 210.6 GB | 84% |
| Q3_K_S | 3.64 | 219.0 GB | 85% |
| IQ3_M | 3.76 | 226.2 GB | 86% |
| Q3_K_M | 4 | 240.6 GB | 88% |
| Q3_K_L | 4.3 | 258.6 GB | 90% |
| IQ4_XS | 4.46 | 268.2 GB | 92% |
| Q4_K_S | 4.67 | 280.8 GB | 93% |
| Q4_K_M | 4.89 | 294.0 GB | 94% |
| Q5_K_S | 5.57 | 334.8 GB | 96% |
| Q5_K_M | 5.7 | 342.6 GB | 96% |
| Q6_K | 6.56 | 394.3 GB | 97% |
| Q8_0 | 8.5 | 510.7 GB | 100% |
| FP16 | 16 | 960.9 GB | 100% |
Benchmark Scores
SWE-bench69.6
How to Run Qwen3-Coder-480B-A35B
Run Qwen3-Coder-480B-A35B locally with Ollama (needs 294.0 GB VRAM at Q4_K_M):
ollama run qwen3:480.2bCompatible GPUs (2)
GPUs that can run Qwen3-Coder-480B-A35B at Q4_K_M quantization: