GPT-5.2 vs Qwen2.5 Coder 32B Instruct
| GPT-5.2 | Qwen2.5 Coder 32B Instruct | |
|---|---|---|
| Coding index | 58.9 | 30.5 |
| Benchmarks behind it | Partial, 2 benchmarks | Partial, 2 benchmarks |
| Output $/Mtok | $14.00 | $1.00 |
| Cost per solved task | $1.91 | Not computable |
| Context window | 400K | 33K |