Model comparison
Qwen3-Coder vs Qwen2.5-Coder
- Input
- $0.22 / 1M tokens
- Output
- $1.80 / 1M tokens
- Context
- 256K (1M with YaRN)
- Released
- 2025-07-22
- Input
- $0.66 / 1M tokens
- Output
- $1.00 / 1M tokens
- Context
- 128K (131,072 tokens) for 7B/14B/32B; 32K for 0.5B/1.5B/3B
- Released
- 2024-11-12
Qwen3-Coder is cheaper on input tokens ($0.2 vs $0.7 per 1M, 3×); Qwen3-Coder has the larger context window (256K vs 128K).
Specs and pricing side by side
| Spec | Qwen3-Coder | Qwen2.5-Coder |
|---|---|---|
| Maker | Alibaba (Qwen) | Alibaba (Qwen)same |
| Released | 2025-07-22 | 2024-11-12 |
| Context window | 256K (1M with YaRN) | 128K (131,072 tokens) for 7B/14B/32B; 32K for 0.5B/1.5B/3B |
| Max output | 65,536 tokens | 8,192 tokens (default generation); hosted… |
| Input price | $0.22 / 1M tokens | $0.66 / 1M tokens |
| Output price | $1.80 / 1M tokens | $1.00 / 1M tokens |
| Parameters | 480B total / 35B active (MoE) | Family of 6 sizes: 0.5B, 1.5B, 3B, 7B, 14B, 32B (flagship 32B = 32.5B total / 31.0B non-embedding) |
| License | Apache 2.0 | Apache 2.0 (all sizes except the 3B model, which uses the Qwen Research license) |
| Weights | Open weights | Open weightssame |
| Modalities | Text | Textsame |
| Knowledge cutoff | not published | Not officially published for Qwen2.5-Coder;… |
Benchmarks both models report
These two models have no benchmark in common in our registry, so there is no like-for-like score to compare. See each model’s own results: Qwen3-Coder, Qwen2.5-Coder.
Which one should you pick?
Pick Qwen3-Coder if…
- Cost matters: input tokens cost 3× less ($0.2 vs $0.7 per 1M).
- You need the longer context (256K vs 128K).
- Your use case is: autonomous agentic coding: multi-step bug fixing, feature implementation, and refactoring inside an agent loop.
Pick Qwen2.5-Coder if…
- Your use case is: self-hosted code assistants and chat-based coding helpers where open weights matter.
FAQ
Which is better, Qwen3-Coder or Qwen2.5-Coder?
Qwen3-Coder is cheaper on input tokens ($0.2 vs $0.7 per 1M, 3×); Qwen3-Coder has the larger context window (256K vs 128K). The better pick depends on whether you value benchmark scores, price per token, context length or openness for your workload.
How much cheaper is Qwen3-Coder?
Qwen3-Coder: $0.22 / 1M tokens input, $1.80 / 1M tokens output. Qwen2.5-Coder: $0.66 / 1M tokens input, $1.00 / 1M tokens output (list API prices as published by each maker).
What are the context windows of Qwen3-Coder and Qwen2.5-Coder?
Qwen3-Coder offers 256K (1M with YaRN) with up to 65,536 tokens output; Qwen2.5-Coder offers 128K (131,072 tokens) for 7B/14B/32B; 32K for 0.5B/1.5B/3B with up to 8,192 tokens (default generation); hosted endpoints commonly cap output at 32,768 output.
Is Qwen3-Coder newer than Qwen2.5-Coder?
Qwen3-Coder was released 2025-07-22; Qwen2.5-Coder was released 2024-11-12.