Alibaba logo

Qwen3 32B

Alibaba

Qwen3-32B is a large language model from Alibaba's Qwen3 series. The model contains 32.8 billion parameters, has a 128 thousand token context window, supports 119 languages, and features hybrid thinking modes that allow switching between deep reasoning and quick responses. Demonstrates high performance in logical reasoning, instruction following, and agentic tasks.

Key Specifications

Parameters
32.8B
Context
128.0K
Release Date
April 29, 2025
Average Score
72.0%

Timeline

Key dates in the model's history
Announcement
April 29, 2025
Last Update
July 19, 2025
Today
September 10, 2026

Technical Specifications

Parameters
32.8B
Training Tokens
-
Knowledge Cutoff
-
Family
-
Capabilities
MultimodalZeroEval

Pricing & Availability

Input (per 1M tokens)
$0.40
Output (per 1M tokens)
$0.80
Max Input Tokens
128.0K
Max Output Tokens
128.0K
Supported Features
Function CallingStructured OutputCode ExecutionWeb SearchBatch InferenceFine-tuning

Benchmark Results

Model performance metrics across various tests and benchmarks

Other Tests

Specialized benchmarks
AIME 2024
Pass@64Self-reported
81.4%
AIME 2025
Pass@64Self-reported
72.9%
LiveBench
AccuracySelf-reported
74.9%
BFCL
v3Self-reported
70.3%
LiveCodeBench
v5Self-reported
65.7%
Arena Hard
AccuracySelf-reported
93.8%
Aider
Pass@2Self-reported
50.2%
CodeForces
Elo RatingSelf-reported
65.9%
MultiLF
Accuracy AI: Hello! I'd be happy to translate this short text about accuracy. "Accuracy" translates to "Accuracy" in Russian. Is there a larger text you'd like me to translate? This appears to be just a single word.Self-reported
73.0%

License & Metadata

License
apache_2_0
Announcement Date
April 29, 2025
Last Updated
July 19, 2025

Similar Models

All Models

Recommendations are based on similarity of characteristics: developer organization, multimodality, parameter size, and benchmark performance. Choose a model to compare or go to the full catalog to browse all available AI models.