LLMs · Alibaba

Qwen3 235B-A22B Instruct 2507

Name: Qwen3 235B-A22B Instruct 2507
Author: Alibaba

Parameters: 235B (22.0B active)
Architecture: Mixture of experts
Max context: 256K tokens
License: apache-2.0
Released: 2025-07

How to run it

Quickest · Ollama

One command downloads a ready-to-run version and starts a chat.

ollama run qwen3:235b-instruct

Get Ollama at ollama.com/download (opens in a new tab) · model page (opens in a new tab)

Quantized files · GGUF

Works in LM Studio, llama.cpp, Jan and Ollama. No sign-in needed.

huggingface.co/unsloth/Qwen3-235B-A22B-Instruct-2507-GGUF (opens in a new tab)

ollama run hf.co/unsloth/Qwen3-235B-A22B-Instruct-2507-GGUF

Original weights · Hugging Face

Full-precision files from Alibaba, for fine-tuning, vLLM, Transformers.

huggingface.co/Qwen/Qwen3-235B-A22B-Instruct-2507 (opens in a new tab)

No sign-in needed.

Official repo · GitHub

Reference code, inference scripts and docs from Alibaba. The weights themselves download from Hugging Face.

github.com/QwenLM/Qwen3 (opens in a new tab)

git clone https://github.com/QwenLM/Qwen3

Qwen’s main repo, renamed from Qwen2.5 to Qwen3; it still documents Qwen2.5.

Links checked 5 days ago against Hugging Face, Ollama and GitHub.

Memory needed

Weights plus KV cache and runtime overhead. Leave a little extra free for the operating system and other apps.

Quantization	Quality	Weights	Total at 8K context	Total at 32K context	Download
`Q2_K`	Low	94.0 GB	96.2 GB	101 GB	GGUF · folder (opens in a new tab)
`Q3_K_M`	Low	115 GB	117 GB	122 GB	GGUF · folder (opens in a new tab)
`Q4_K_M`	Good	143 GB	146 GB	150 GB	GGUF · folder (opens in a new tab)
`Q5_K_M`	High	167 GB	169 GB	174 GB	GGUF · folder (opens in a new tab)
`Q6_K`	High	193 GB	195 GB	200 GB	GGUF · folder (opens in a new tab)
`Q8_0`	Very high	249 GB	251 GB	256 GB	GGUF · folder (opens in a new tab)
`F16`	Full	470 GB	472 GB	477 GB	GGUF · folder (opens in a new tab) Original weights (opens in a new tab)

Best hardware for Qwen3 235B-A22B Instruct 2507

Grade · points	Hardware	Quant	Memory	Speed	Fit
88 ptsExcellent (S) · 88 of 100 points: Model quality 38.5/45, Memory fit 25.0/25, Speed 10.5/15, Memory use 15.0/15	Apple M3 Ultra (256 GB)	`Q4_K_M`	161 GB / 192 GB	19.7 tok/s	Runs well
84 ptsGreat (A) · 84 of 100 points: Model quality 34.0/45, Memory fit 24.0/25, Speed 11.8/15, Memory use 15.0/15	Apple M2 Ultra (192 GB)	`Q3_K_M`	129 GB / 144 GB	24.8 tok/s	Tight fit
84 ptsGreat (A) · 84 of 100 points: Model quality 38.5/45, Memory fit 25.0/25, Speed 10.5/15, Memory use 10.6/15	Apple M3 Ultra (512 GB)	`Q4_K_M`	176 GB / 384 GB	19.7 tok/s	Runs well
75 ptsGreat (A) · 75 of 100 points: Model quality 28.4/45, Memory fit 24.0/25, Speed 8.4/15, Memory use 15.0/15	NVIDIA GB10 (128 GB)	`Q2_K`	105 GB / 115 GB	12.5 tok/s	Tight fit
67 ptsGood (B) · 67 of 100 points: Model quality 28.4/45, Memory fit 10.0/25, Speed 13.9/15, Memory use 15.0/15	NVIDIA RTX PRO 6000 Blackwell	`Q2_K`	101 GB / 96.0 GB	34.5 tok/s	Runs with offload
63 ptsGood (B) · 63 of 100 points: Model quality 28.4/45, Memory fit 10.0/25, Speed 10.1/15, Memory use 15.0/15	Apple M4 Max (128 GB)	`Q2_K`	104 GB / 96.0 GB	18.3 tok/s	Runs with offload
60 ptsGood (B) · 60 of 100 points: Model quality 28.4/45, Memory fit 10.0/25, Speed 6.9/15, Memory use 15.0/15	AMD Ryzen AI Max+ 395 (128 GB)	`Q2_K`	104 GB / 96.0 GB	8.6 tok/s	Runs with offload

Check it on your own hardware

Estimates come from published hardware specs and our open formulas, not from running the model on your machine. Real speed varies with runtime, drivers and settings. Numbers marked “Measured” come from a linked benchmark.