/ Models Index / Qwen 2.5 Coder

Qwen 2.5 Coder

Specialty · Code · Alibaba · Released Q3 2024
B+

Good (B+) — Capability Grade

Qwen 2.5 Coder is Alibaba's code-specialized open-weights model. Strong HumanEval and SWE-Bench scores at 32B parameter scale. Apache 2.0 licensed.

83
Composite / 100
/ Subscore Breakdown · 6 Capability Dimensions

Where this grade comes from.

General Reasoning
B
Code Generation
A-
Math & STEM
B
Tool Use & Agency
B
Multimodal
D
Safety & Alignment
B
/ Key Events & Disclosures

Release timeline & positioning.

  • Released Q3 2024
  • Code-specialized
  • Apache 2.0
  • Strong SWE-Bench

/ Best for

Code generation workloads requiring on-premise / open-weights deployment.

/ Watch out for

Code-specialty; not suitable for general reasoning workloads.