Qwen3.5 397B-A17B scores 27.6 on the Respan Index (#72), ahead of GPT-5.2-Codex at 26.4 (#74). Qwen3.5 397B-A17B leads on 1 of the 1 benchmarks both report. Qwen3.5 397B-A17B is 3.6x cheaper per token at list price. GPT-5.2-Codex has the larger context window (400K tokens).
Higher is better. Each score links to where it was published.
* Estimated: no published scores in that category. How the Respan Index works