Skip to main content
leaderboard
.cn
Toggle Sidebar
Search models or organizations…
⌘K
Explore
Frontier Models
Historical Trends
Local Models
Compare Models
Ranking Studio
Library
Models
Pricing
Organizations
Benchmarks
Data sources
Settings
Toggle Sidebar
leaderboard
.cn
Loading page content
Feedback
Loading page content
Back to models
Qwen3-235B-A22B
Open
Share
Add to comparison
Tool use · Reasoning
Overview
Deployment
Evaluations
Evaluation overview
Overall rank
#100
69.3 points · 202 models
Open-weight rank
#38
72.3 points · 120 models
Data coverage
110 items
3 sources · updated through 2026-09-02
Domain performance
Coding
#95 / 182
67
Research
#97 / 196
78
Agents
#101 / 148
63
Long context
#68 / 95
70.5
Evaluation highlights
All evaluations
LiveCodeBench v5
Coding · Qwen3 Technical Report
70.7%
1
/ 36
LiveBench 20241125
Reasoning · Qwen3 Technical Report
77.1%
3
/ 42
RULER 128K
Long context · Qwen3 Technical Report · Non-reasoning mode
90.6%
1
/ 10
BFCL v3 Full
Agents · Qwen3 Technical Report
70.8%
9
/ 49
MLogiQA
Multilingual · Qwen3 Technical Report
77.1%
1
/ 35
Arena-Hard
Conversation · Qwen3 Technical Report · Non-reasoning mode
96.1%
2
/ 39
Basic information
Context (input / output)
262K / 33K
Parameters
235B / 22B active
Release date
2025-04
License
Apache 2.0
Input modalities
Text
Output modalities
Text
Architecture
MoE
Modes
Supports thinking and non-thinking modes
Version
Initial release, April 2025
Availability
vLLM · FP8
Deployment options
FP8 · 4 × 141 GB VRAM
Related links
Official site
HuggingFace
Version relationships
Qwen3-235B-A22B family
Qwen3-235B-A22B
Current
2025-04 · Current model
Qwen3-235B-A22B-Base
2025-04 · Same series
Qwen3-235B-A22B-Instruct-2507
2025-07 · Same series
Qwen3-235B-A22B-Thinking-2507
2025-07 · Same series