Skip to main content

Qwen2.5 32B Instruct

Alibaba Cloud / Qwen TeamQwenOpen WeightApache 2.0 · Commercial OK

Description

Qwen2.5-32B-Instruct is an instruction-tuned 32 billion parameter language model, part of the Qwen2.5 series. It is designed to follow instructions, generate long texts (over 8K tokens), understand structured data (e.g., tables), and generate structured outputs, especially JSON. The model supports multilingual capabilities across over 29 languages.

Release Date
2024-09-19
Parameters
32.5B
Context Length
Modalities

Capability Radar

70
general
90
coding
80
reasoning
43
scienceest.
0
agents
0
multimodal

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

Rankings

Domain#RankScoreSource
Reasoning48
69.0
LS

Benchmark Scores (LLM Stats)

Biology

GPQA49.5%SR

Chemistry

MMLU-STEM80.9%SR

Code

HumanEval88.4%SR

Finance

MMLU83.3%SR
MMLU-Pro69.0%SR
TruthfulQA57.8%SR
TheoremQA44.1%SR

General

MBPP0.84 / 100SR
MMLU-Redux83.9%SR
MultiPL-E75.4%SR
ARC-C70.4%SR
MBPP+67.2%SR

Language

BBH84.5%SR
Winogrande82.0%SR

Math

GSM8k95.9%SR
MATH83.1%SR

Reasoning

HellaSwag85.2%SR
HumanEval+52.4%SR

AA Evaluation Indices

No AA evaluation data available

LLM Stats Category Scores

Code
90
Language
80
Math
80
General
70
Healthcare
70
Legal
70
Reasoning
70
Finance
60
Biology
50
Chemistry
50
Physics
50

Pricing

No pricing data available

Speed

No speed data available

Available Providers

(LS internal units)

No provider data available

External Sources