Qwen3-Coder-Next
Qwen3-Coder-Next is an open-weights language model from Qwen, released February 2, 2026. It has 80B parameters (mixture-of-experts, 3.9B active). The context window is 256K tokens.
Today, we're announcing Qwen3-Coder-Next, an open-weight language model designed specifically for coding agents and local development. It features the following key enhancements: Super Efficient with Significant Performance: With only 3B activated parameters (80B total parameters), it achieves performance comparable …
Check it against your own PC →
Can you run Qwen3-Coder-Next locally?
The smallest common GPU it runs on is 6 GB of VRAM (with 32 GB of RAM): IQ2_M, GPU + experts in RAM, roughly 40 tokens per second. Estimates assume 32 GB of DDR5 RAM, a 16K context and llama.cpp-style GGUF quantization. Enter your exact hardware in the live app.
| GPU memory | Verdict | Best quant | VRAM | RAM | tok/s |
|---|---|---|---|---|---|
| 6 GB VRAM | Runs · GPU + CPU experts | IQ2_M (heavy loss) | 5.2 GB | 22.7 GB | ~40 |
| 8 GB VRAM | Runs · GPU + CPU experts | IQ3_XXS (noticeable loss) | 7.2 GB | 24.3 GB | ~40 |
| 12 GB VRAM | Runs · GPU + CPU experts | IQ3_XXS (noticeable loss) | 11.2 GB | 20.3 GB | ~47 |
| 16 GB VRAM | Runs · GPU + CPU experts | Q3_K_M (okay) | 15.2 GB | 24.7 GB | ~42 |
| 24 GB VRAM | Runs · GPU + CPU experts | IQ4_XS (good) | 23.2 GB | 20.1 GB | ~52 |
| 32 GB VRAM | Runs · GPU + CPU experts | Q4_K_M (very good) | 31.2 GB | 18.5 GB | ~59 |
Ready-made quantizations on Hugging Face: unsloth/Qwen3-Coder-Next-GGUF · lmstudio-community/Qwen3-Coder-Next-GGUF · bartowski/Qwen_Qwen3-Coder-Next-GGUF · ggml-org/Qwen3-Coder-Next-GGUF · MaziyarPanahi/Qwen3-Coder-Next-GGUF · qtum/Qwen3-Coder-Next-AWQ
Benchmark results
Epoch AI has not published results for Qwen3-Coder-Next yet. New models are usually evaluated within days to weeks.
Links
Announcement · Hugging Face: Qwen/Qwen3-Coder-Next
More from Qwen
Qwen3.8 Omni Flash · Qwen3.8 Max (0902) · Qwen3.8 Flash · Qwen3.8 Flash Next · Qwen 3.8 27B · Qwen3.8-2.4T-A95B · Qwen 3.8 Max · Qwen3.7 Flash