Skip to main content
AITier

Qwen3 32B (Non-reasoning)

AlibabaQwenOpen WeightApache 2.0 · Commercial OK

Description

Qwen3-32B is a large language model from Alibaba's Qwen3 series. It features 32.8 billion parameters, a 128k token context window, support for 119 languages, and hybrid thinking modes allowing switching between deep reasoning and fast responses. It demonstrates strong performance in reasoning, instruction-following, and agent capabilities.

Release Date
2025-04-28
Parameters
32.8B
Context Length
41K
Modalities
text

Capability Radar

30
general
29
coding
36
reasoning
35
scienceest.
70
agents
0
multimodal

Science uses a reasoning proxy when dedicated science benchmarks are unavailable.

Rankings

Domain#RankScoreSource
Code Ranking380
16.0
AA
General Ranking311
36.0
AA
Math Reasoning243
36.0
AA
Science304
36.0
AA

Benchmark Scores (LLM Stats)

Code

LiveCodeBench65.7%SR
Aider50.2%SR

Creativity

Arena Hard93.8%SR

General

LiveBench74.9%SR
MultiLF73.0%SR
BFCL70.3%SR

Math

AIME 202481.4%SR
AIME 202572.9%SR
CodeForces0.66 / 3000SR

AA Evaluation Indices

Math Index
19.7
Intelligence Index
14.5
Math 500
0.9
Mmlu Pro
0.7
Gpqa
0.5
Ifbench
0.3
Aime
0.3
Livecodebench
0.3
Scicode
0.3
Aime 25
0.2
Hle
0.0
Lcr
0.0

LLM Stats Category Scores

Writing
90
Creativity
90
General
80
Tool Calling
70
Math
70
Reasoning
70
Code
60

Pricing

Input Price$0.15 / 1M tokens
Output Price$0.59 / 1M tokens
Blended Price (3:1)$0.26 / 1M tokens

Speed

Tokens/sec89.5 tokens/s
Time to First Token1.14s
Time to Answer1.14s

Available Providers

(LS internal units)
ProviderInput PriceOutput Price
DeepInfra100K300K
Novita100K440K
Sambanova400K800K

External Sources

LLM Stats