GPT-5.6 Terra vs Qwen2.5 Coder 32B Instruct
| GPT-5.6 Terra | Qwen2.5 Coder 32B Instruct | |
|---|---|---|
| Coding index | 69.0 | 30.5 |
| Benchmarks behind it | Partial, 2 benchmarks | Partial, 2 benchmarks |
| Output $/Mtok | $12.00 | $1.00 |
| Cost per solved task | $0.42 | Not computable |
| Context window | 1.05M | 33K |