NEWHow AI models draw the world map · Gecko Tests
Beta
Home/Models/Qwen2.5 Coder 32B Instruct
Alibaba Qwen logo

Qwen2.5 Coder 32B Instruct

by Alibaba Qwen · Released Nov 2024

Open Source
71.4
avg score
Rank #34
Compare
Better than 89% of all models
Context
33K tokens (~16 books)
Input $/1M
$0.66
Output $/1M
$1.00
Type
text
License
Open Source
Benchmarks
14 tested
Data as of
About

Qwen2.5-Coder is the latest series of Code-Specific Qwen large language models (formerly known as CodeQwen). Qwen2.5-Coder brings the following improvements upon CodeQwen1.5: - Significantly improvements in **code generation**, **code reasoning**...

Tested on 14 benchmarks · BenchGecko score 71.4. Top scores: Chatbot Arena Elo — Overall (1270.6%), GSM8K (91.1%), HellaSwag (77.3%).

Looking for similar performance at lower cost?
MiniMax M2 scores 72.4 (101% as good) at $0.30/1M input · 55% cheaper
Capabilities
coding
43.9
#123 globally
reasoning
13.7
#154 globally
math
70.3
#43 globally
knowledge
53.8
#94 globally
general
52.3
#27 globally
language
72.7
#74 globally
Benchmark Scores
Compare All
Tested on 14 benchmarks · Ranked across 7 categories
Score Distribution (all 312 models)
0255075100
▲ You are here
codingCompare coding →
Aider — Code Editing

Code editing benchmark from the Aider project. Measures ability to apply targeted code changes while maintaining correctness and style.

71.4·
Aider polyglot

Multi-language code editing from Aider. Tests editing ability across Python, JavaScript, TypeScript, Java, C++, Go, Rust, and more.

16.4·
reasoningCompare reasoning →
MUSR

HuggingFace MuSR (Multi-Step Reasoning). Tests multi-hop reasoning requiring chaining multiple facts together.

13.7·
mathCompare math →
GSM8K

Grade school math word problems. 8,500 problems testing multi-step arithmetic reasoning. A foundational math benchmark.

91.1·
MATH Level 5

HuggingFace evaluation of MATH Level 5 problems. Competition math requiring advanced reasoning and proof construction.

49.5·
Quick compare:Anthropic logovs Claude Opus 4.8Alibaba logovs Qwen2.5 14B Instructz-ai logovs GLM 4.5
Excellent (85+) Good (70-85) Average (50-70) Below (<50)
Model Family · Alibaba Qwen Qwen 2.5
Sep 2024
59.6
$0.36/M in33Kctx28 benchmarks
Oct 2024
34.8-24.8
$0.10/M in(-0.26)33Kctx13 benchmarks
Nov 2024
71.4+36.6
$0.66/M in(+0.56)33Kctx14 benchmarks
Apr 2025
56.3-15.1
$0.03/M in(-0.63)33Kctx12 benchmarks
Mar 2025
$0.20/M in(+0.17)128Kctx(+95K)
Feb 2025
$0.80/M in(+0.60)128Kctx
Jan 2024
54.6-1.7
n/an/actx4 benchmarks
Jan 2024
36.2-18.4
n/an/actx8 benchmarks
See the full Qwen 2.5 family →
Similar Models
Anthropic logo
Claude Opus 4.8
Anthropic
71.8$5.00/1M
Alibaba logo
Qwen2.5 14B Instruct
Alibaba
71.0n/a
z-ai logo
GLM 4.5
z-ai
72.3$0.60/1M
Links
Info
Alibaba QwenPricing explorerDevelopers · API
Research
Technical Report
Documentation
API DocsPlayground
Community
@Alibaba Qwen
Source Code
GitHubHugging Face
BenchGecko API
qwen-2-5-coder-32b-instruct
Specifications
  • Typetext
  • Context33K tokens (~16 books)
  • ReleasedNov 2024
  • LicenseOpen Source
  • StatusActive
  • Cost / Message~$0.002
Available On
Alibaba Qwen logoAlibaba Qwen$0.66
Categories
codingreasoningmathknowledgegenerallanguage
Learn More
transformeropen weightstokens
Share & Export
Tweet
Related Models
Anthropic logoClaude Opus 4.8
71.8
Alibaba logoQwen2.5 14B Instruct
71.0
z-ai logoGLM 4.5
72.3
minimax logoMiniMax M2
72.4
Qwen2.5 Coder 32B Instruct is an open-source text AI model by Alibaba Qwen, released in November 2024. It has an average benchmark score of 71.4. Context window: 33K tokens.

Related Models

Claude Opus 4.8 · AnthropicQwen2.5 14B Instruct · AlibabaGLM 4.5 · z-aiMiniMax M2 · minimaxGPT-5.2 Pro · OpenAI

Benchmarks

Chatbot Arena Elo — OverallGSM8KHellaSwagIFEvalMMLU

Related Pages

Alibaba Qwen · ProviderAlibaba Qwen · EconomyAll ModelsCompare ModelsPricingDevelopers · API

Key facts · as of 2026-10-05

  • Qwen2.5 Coder 32B Instruct by Alibaba Qwen. BenchGecko score 71.4, rank 34 of 312 scored models (normalized average of public benchmark scores).
  • List price $0.66 input · $1.00 output per 1M tokens (as of 2026-10-05).
  • Sold by 1 provider (as of 2026-10-05): Cloudflare $0.66 in / $1.00 out.

How to cite · data as of 2026-10-05

Qwen2.5 Coder 32B Instruct · benchmarks, pricing and providers. BenchGecko, data as of 2026-10-05. https://benchgecko.ai/model/qwen-2-5-coder-32b-instruct

Credit "Source: BenchGecko" with a link. Prices per provider and Gecko Tests are BenchGecko data (CC BY 4.0); benchmark scores keep their original source, listed in the JSON. JSON · llms.txt · MCP