Skip to content

Qwen3 Next 80B A3B Instruct

Qwen
Code Multilingual Tool Calls

Qwen3 Next 80B A3B Instruct is a Mixture-of-Experts model from Alibaba's Qwen team with 81.32 billion total parameters, fine-tuned for instruction following and tool-use workflows. Only around 3 billion parameters activate per token, activating 10 of 512 experts, matching the performance of much larger models at dramatically lower compute cost. It supports code generation, tool calling, and 13 languages including English and Chinese. With a 262K context window and flash attention, it processes long documents natively and quantizes well to GGUF for self-hosted inference on consumer-grade multi-GPU setups.

Hardware Configuration

Optional — for precise deployment recommendations
Quantization Quality Size Fit
Q8_0 High 78.99 GB
Q8_K_XL High 86.69 GB
Q6_K High 61.04 GB
Q6_K_XL High 63.81 GB
Q5_K_M Medium 52.91 GB
Q5_K_S Medium 51.24 GB
Q5_K_XL Medium 52.77 GB
Q4_K_M Medium 45.17 GB
Q4_K_S Medium 42.38 GB
Q4_K_XL Medium 42.9 GB
Q4_0 Medium 42.2 GB
Q4_1 Medium 46.61 GB
Q3_K_M Low 35.67 GB
Q3_K_S Low 32.21 GB
Q3_K_XL Low 33.19 GB
Q2_K Low 27.17 GB
Q2_K_L Low 27.24 GB
Q2_K_XL Low 28.06 GB
Last updated: March 5, 2026