24.4pp above Opus 4.6 (53.4%). Massive jump on harder coding eval.

#1 open source, #3 globally. Beats GPT-5.4 (57.7) and Claude Opus 4.6 (57.3). First Chinese model to top SWE-bench Pro. Zero Nvidia hardware.

Harder coding benchmark variant. GPT-5.4 splits coding, reasoning, and computer use into one unified model.

Leads SWE-bench Pro. 25% faster than GPT-5.2 Codex. Covers 1,865 multi-language tasks.

Matches GPT-5.3 Codex level. Nearly approaches Opus best on SWE-Pro.

Smallest GPT-5.4 variant. 2x faster than GPT-5 mini at similar coding quality.

Competitive with GPT-5.4 (57.7%). Behind Claude Mythos Preview (77.8%).

Fastest, cheapest GPT-5.4 variant. Designed for classification, data extraction, and coding subagents.

Trails Claude Mythos (77.8%) significantly but competitive with other frontier models.

Internal evaluation framework with minimal tool set. 1T param MoE, 32B active.

Flagship coding model from ByteDance Seed team.

Nearly matches Pro variant (46.9) on harder SWE subset.

Competitive with much larger models on harder SWE subset.