Skip to main content
brand
context
industry
strategy
AaaS
Modelllm-modelsv1.0

Qwen 2.5 Coder

by Alibaba Cloud · open-source · Last verified 2026-04-24

Qwen 2.5 Coder is Alibaba's code-specialized model achieving performance comparable to GPT-4o on HumanEval and SWE-bench at 32B parameters. Available as open weights, it supports 92 programming languages and is widely adopted for code completion, debugging, and code review agent workflows.

https://huggingface.co/Qwen/Qwen2.5-Coder-32B-Instruct
C
CBelow Average
Adoption: C+Quality: B+Freshness: ACitations: CEngagement: F

Specifications

License
Open Source
Pricing
open-source
Capabilities
Integrations
Use Cases
API Available
No
Modalities
Tags
qwen, coding, open-weights, humaneval, swe-bench, code-completion, alibaba
Added
2026-04-24
Completeness
60%

Index Score

44
Adoption
50
Quality
70
Freshness
80
Citations
40
Engagement
0

Need help choosing the right model?

Get Expert Guidance

Explore the full AI ecosystem on Agents as a Service