Across 13 shared benchmarks, Kimi K2.5 scores higher on 7 and nvidia-nemotron-3-ultra-550b-a55b on 6. The widest gap is HLE, where Kimi K2.5 scores 50.2 against 26.7. Tracked API pricing per million tokens: Kimi K2.5 $0.56 in / $2.94 out, nvidia-nemotron-3-ultra-550b-a55b $0.60 in / $2.75 out.
| Benchmark | Kimi K2.5 | nvidia-nemotron-3-ultra-550b-a55b |
|---|---|---|
| browsecomp | 74.9 | 44.4 |
| gdpval | 38.3 | 46.7 |
| GPQA Diamond | 87.9 | 87 |
| HLE | 50.2 | 26.7 |
| HLE (with tools) | 51.8 | 37.4 |
| imo_answer_bench | 81.8 | 92.3 |
| LiveCodeBench v6 | 85 | 89 |
| longbench_v2 | 61 | 61.9 |
| MMLU-Pro | 87.1 | 86.8 |
| multichallenge | 61.4 | 63.8 |
| SWE-bench Multilingual | 73 | 67.7 |
| SWE-bench Verified | 76.8 | 70.7 |
| Terminal-Bench 2.1 | 45.7 | 56.4 |
Best tracked score per model per benchmark (default configuration; source-attributed). ↓ marks lower-is-better metrics. Open either model for its full surface, provenance and pricing. Quoted rates are the price-setter row we currently track for each model — its direct or vendor-official listing where one exists (moonshot-official, direct), otherwise the lowest tracked offer.