LLM EXPLORER
 LLM Explorer: A Curated Large Language Model Directory and Analytics
Dark Theme
  Agents   Benchmarks   LLM  List    LLM Hosting    LLM Leaderboards    Blog    Newsfeed    Advertise

Qwen3 235B A22B Thinking 2507 by Qwen

 »  All LLMs  »  Qwen  »  Qwen3 235B A22B Thinking 2507   URL Share it on

  37143   

Qwen3 235B A22B Thinking 2507 is an open-source language model by Qwen. Features: 235b LLM, VRAM: 171.9GB, Context: 256K, License: apache-2.0, LLM Explorer Score: 0.31.

  Arxiv:2402.17463   Arxiv:2404.06654   Arxiv:2407.02490   Arxiv:2501.15383   Arxiv:2505.09388   Conversational   Deploy:azure   Endpoints compatible   Eval-results   Qwen3 moe   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/Qwen/Qwen3-235B-A22B-Thinking-2507 

Qwen3 235B A22B Thinking 2507 Benchmarks

LLME Score: 0.31485
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Qwen3 235B A22B Thinking 2507 (Qwen/Qwen3-235B-A22B-Thinking-2507)
🌟 Advertise your project 🚀

Qwen3 235B A22B Thinking 2507 Parameters and Internals

LLM NameQwen3 235B A22B Thinking 2507
Repository 🤗https://huggingface.co/Qwen/Qwen3-235B-A22B-Thinking-2507 
Model Size235b
Required VRAM171.9 GB
Updated2026-05-07
MaintainerQwen
Model Typeqwen3_moe
Model Files  4.0 GB: 1-of-118   4.0 GB: 2-of-118   4.0 GB: 3-of-118   4.0 GB: 4-of-118   4.0 GB: 5-of-118   4.0 GB: 6-of-118   4.0 GB: 7-of-118   4.0 GB: 8-of-118   4.0 GB: 9-of-118   4.0 GB: 10-of-118   4.0 GB: 11-of-118   4.0 GB: 12-of-118   4.0 GB: 13-of-118   4.0 GB: 14-of-118   4.0 GB: 15-of-118   4.0 GB: 16-of-118   4.0 GB: 17-of-118   4.0 GB: 18-of-118   4.0 GB: 19-of-118   4.0 GB: 20-of-118   4.0 GB: 21-of-118   4.0 GB: 22-of-118   4.0 GB: 23-of-118   3.9 GB: 24-of-118   4.0 GB: 25-of-118   4.0 GB: 26-of-118   4.0 GB: 27-of-118   4.0 GB: 28-of-118   4.0 GB: 29-of-118   4.0 GB: 30-of-118   4.0 GB: 31-of-118   4.0 GB: 32-of-118   4.0 GB: 33-of-118   4.0 GB: 34-of-118   4.0 GB: 35-of-118   4.0 GB: 36-of-118   4.0 GB: 37-of-118   4.0 GB: 38-of-118   4.0 GB: 39-of-118   4.0 GB: 40-of-118   4.0 GB: 41-of-118   4.0 GB: 42-of-118   4.0 GB: 43-of-118
Model ArchitectureQwen3MoeForCausalLM
Licenseapache-2.0
Context Length262144
Model Max Length262144
Transformers Version4.51.0
Vocabulary Size151936
Torch Data Typebfloat16

Quantized Models of the Qwen3 235B A22B Thinking 2507

Model
Likes
Downloads
VRAM
...3 235B A22B Thinking 2507 8bit3250227 GB
...5B A22B Thinking 2507 3bit DWQ6197103 GB
...5B A22B Thinking 2507 4bit DWQ489131 GB

Best Alternatives to Qwen3 235B A22B Thinking 2507

Best Alternatives
Context / RAM
Downloads
Likes
...en3 235B A22B Instruct 2507 1M986K / 171.9 GB31
Qwen3 235B A22B Instruct 2507256K / 171.9 GB135732782
Qwen3 235B A22B Instruct 2507256K / 167.9 GB919
Qwen3 235B A22B Thinking 2507256K / 167.9 GB142
MiroThinker 1.7256K / 220 GB1350138
...n3 235B A22B Instruct 2507 FP8256K / 236.5 GB353515147
MiroThinker 1.7 Mini256K / 61.1 GB133799
... 235B A22B Thinking 2507 NVFP4256K / 139.2 GB228288
...n3 235B A22B Thinking 2507 FP8256K / 236.5 GB3081284
... Nemotron 235B A22B GenRM 2603256K / 195 GB31327
Note: green Score (e.g. "73.2") means that the model is better than Qwen/Qwen3-235B-A22B-Thinking-2507.

Rank the Qwen3 235B A22B Thinking 2507 Capabilities

🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟

Instruction Following and Task Automation  
Factuality and Completeness of Knowledge  
Censorship and Alignment  
Data Analysis and Insight Generation  
Text Generation  
Text Summarization and Feature Extraction  
Code Generation  
Multi-Language Support and Translation  

What open-source LLMs or SLMs are you in search of? 54415 in total.

Email us: info@extractum.io. Our Privacy Policy  | Terms and Conditions  | Suggest an improvement.
Our Social Media →  
Original data from HuggingFace, OpenCompass and various public git repos.
Check out Ag3ntum — our secure, self-hosted AI agent for server management.
Release v20260328a
  Support LLM Explorer