Skip to main content

QwQ 32B

AlibabaQwenOpen WeightApache 2.0 · Commercial OK

Description

A model focused on advancing AI reasoning capabilities, particularly excelling in mathematics and programming. Features deep introspection and self-questioning abilities while having some limitations in language mixing and recursive/endless reasoning patterns.

Release Date
2025-03-05
Parameters
32.5B
Context Length
Modalities

Capability Radar

35
general
57
coding
54
reasoning
41
scienceest.
70
agents
0
multimodal

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

Rankings

Domain#RankScoreSource
Code Ranking147
51.0
AA
General Ranking236
44.0
AA
Math Reasoning171
53.0
AA
Science233
45.0
AA

Benchmark Scores (LLM Stats)

Biology

GPQA65.2%SR

Code

LiveCodeBench63.4%SR

General

IFEval83.9%SR
LiveBench73.1%SR
BFCL66.4%SR

Math

MATH-50090.6%SR
AIME 202479.5%SR

AA Evaluation Indices

Math Index
29.0
Intelligence Index
19.7
Math 500
1.0
Aime
0.8
Mmlu Pro
0.8
Livecodebench
0.6
Gpqa
0.6
Ifbench
0.4
Scicode
0.4
Aime 25
0.3
Lcr
0.3
Hle
0.1

LLM Stats Category Scores

Structured Output
80
Instruction Following
80
Math
80
Tool Calling
70
Biology
70
Chemistry
70
General
70
Physics
70
Reasoning
70
Code
60

Pricing

Input Price$0.66 / 1M tokens
Output Price$1 / 1M tokens
Blended Price (3:1)$0.745 / 1M tokens

Speed

Tokens/sec31.7 tokens/s
Time to First Token0.40s
Time to Answer79.07s

Available Providers

(LS internal units)

No provider data available

External Sources