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
Qwen3-30B-A3B-Thinking-2507
Open
Share
Add to comparison
Tool use · Reasoning
Overview
Deployment
Evaluations
Evaluation overview
Overall rank
#90
67.8 points · 180 models
Open-weight rank
#29
75.1 points · 109 models
Data coverage
24 items
1 source · updated through 2025-07-01
Domain performance
Coding
#80 / 160
69.3
Research
#90 / 174
77.6
Agents
#74 / 125
65.2
Evaluation highlights
All evaluations
LiveBench 20241125
Reasoning · Qwen3 2507 Hugging Face Model Card
76.8%
4
/ 42
BFCL v3 Full
Agents · Qwen3 2507 Hugging Face Model Card
72.4%
5
/ 49
IFEval
Instruction following · Qwen3 2507 Hugging Face Model Card
88.9%
13
/ 30
CFEval
Coding · Qwen3 2507 Hugging Face Model Card
2,044score
2
/ 3
PolyMath
Multilingual · Qwen3 2507 Hugging Face Model Card
52.6%
3
/ 44
WritingBench
Writing · Qwen3 2507 Hugging Face Model Card
85%
6
/ 47
Basic information
Context (input / output)
262K / 33K
Parameters
30.5B / 3.3B active
Release date
2025-07
License
Apache 2.0
Input modalities
Text
Output modalities
Text
Modes
Thinking only
Version
2507
Active parameters
3.3B
Availability
SGLang · FP8 · 125K
Deployment options
FP8
Related links
Official site
HuggingFace
Version relationships
Qwen3-30B-A3B family
Qwen3-30B-A3B
2025-04 · Same series
Qwen3-30B-A3B-Base
2025-04 · Same series
Qwen3-30B-A3B-Instruct-2507
2025-07 · Same series
Qwen3-30B-A3B-Thinking-2507
Current
2025-07 · Current model