bartowski / text-generation updated 6 months ago

Qwen_Qwen3-Next-80B-A3B-Thinking-GGUF

Note: if it's a newly supported model, you may need to wait for an update from the developers.

Params
Context
Downloads 30d
208 K
Likes
16
Commercial use: allowed apache-2.0 Not gated GGUF View on Hugging Face ↗

Download history

daily snapshots · 27 days
279 K184 K
Aug 26Sep 4Sep 13Sep 21

Can you run it?

Estimated VRAM at 8K context unless noted. Pick your hardware to see the verdict per quantization.

FileQuantSizeEst. VRAMVerdict on RTX 4090 · 24 GB
Qwen_Qwen3-Next-80B-A3B-Thinking-IQ1_S.gguf IQ1_S 16.6 GB 18.7 GB ✅ Runs comfortably
Qwen_Qwen3-Next-80B-A3B-Thinking-IQ1_M.gguf IQ1_M 17.3 GB 19.5 GB ✅ Runs comfortably
Qwen_Qwen3-Next-80B-A3B-Thinking-IQ2_XXS.gguf IQ2_XXS 19.3 GB 21.7 GB ⚠️ Tight — reduce context
Qwen_Qwen3-Next-80B-A3B-Thinking-IQ2_XS.gguf IQ2_XS 22.2 GB 24.9 GB ❌ Won’t fit
Qwen_Qwen3-Next-80B-A3B-Thinking-IQ2_S.gguf IQ2_S 23.4 GB 26.2 GB ❌ Won’t fit
Qwen_Qwen3-Next-80B-A3B-Thinking-IQ2_M.gguf IQ2_M 26.1 GB 29.2 GB ❌ Won’t fit
Qwen_Qwen3-Next-80B-A3B-Thinking-IQ3_XS.gguf IQ3_XS 33.0 GB 36.8 GB ❌ Won’t fit
Qwen_Qwen3-Next-80B-A3B-Thinking-IQ3_M.gguf IQ3_M 36.6 GB 40.8 GB ❌ Won’t fit
Estimate: file size × 1.1 + KV cache at 8K + 0.5 GB overhead. Not a benchmark — how we calculate this.

Run it

copy-paste, exact tags checked against the Hub
~ · ollama · IQ1_M
$ ollama run qwen-qwen3-next-80b-a3b-thinking-gguf

# pin the quantization explicitly
$ ollama run qwen-qwen3-next-80b-a3b-thinking-gguf-iq1_m
est. VRAM 19.5 GBon RTX 4090 · 24 GBJSON API →

Specifications

Licence
apache-2.0
First seen on the Hub
2025-09-09
Training datasets
undisclosed
Added to our catalog
2026-08-26
Compare with any text-generation model