Coding · Benchmark profile
LiveCodeBench v6 is a named release slice used in provider comparison tables. Keeping it separate prevents v6 results from being mixed into older or rolling LiveCodeBench windows.
Data verified 27 Jul 2026 · Methodology 1.8.0
Benchmark score on LiveCodeBench v6
Sakana Fugu-Ultra leads at 93.2%, followed by Sakana Fugu (92.9%) and Kimi K2.6 (89.6%).
10 ranked models · higher is better · labels show rank and score
| Model | Rank | Provider | Score |
|---|---|---|---|
| Sakana Fugu-Ultra | #1 | Sakana AI | 93.2% |
| Sakana Fugu | #2 | Sakana AI | 92.9% |
| Kimi K2.6 | #3 | Moonshot AI | 89.6% |
| Qwen3.6 Plus | #4 | Alibaba Cloud | 87.1% |
| Kimi K2.5 | #5 | Moonshot AI | 85% |
| Claude Opus 4.5 | #6 | Anthropic | 84.8% |
| Qwen3.5 397B | #7 | Alibaba Cloud | 83.6% |
| ZAYA1-8B | #8 | Zyphra | 65.8% |
| ZAYA1-74B-Preview | #9 | Zyphra | 65.7% |
| MiniCPM5-1B | #10 | OpenBMB | 33.5% |
One best score per model · higher is better
| Rank | Model | Provider | License | Evidence use | Score |
|---|---|---|---|---|---|
| #1 | Sakana Fugu-Ultra sakana-fugu-ultra | Sakana AI | closed | Estimated reference | 93.2% |
| #2 | Sakana Fugu sakana-fugu | Sakana AI | closed | Estimated reference | 92.9% |
| #3 | Kimi K2.6 kimi-k2.6 | Moonshot AI | open | Estimated reference | 89.6% |
| #4 | Qwen3.6 Plus qwen3-6-plus | Alibaba Cloud | closed | Reference only | 87.1% |
| #5 | Kimi K2.5 kimi-k2.5 | Moonshot AI | open | Reference only | 85% |
| #6 | Claude Opus 4.5 claude-opus-4-5 | Anthropic | closed | Reference only | 84.8% |
| #7 | Qwen3.5 397B qwen3.5-397b-a17b | Alibaba Cloud | open | Estimated reference | 83.6% |
| #8 | ZAYA1-8B zaya1-8b | Zyphra | open | Estimated reference | 65.8% |
| #9 | ZAYA1-74B-Preview zaya1-74b-preview | Zyphra | open | Estimated reference | 65.7% |
| #10 | MiniCPM5-1B minicpm5-1b | OpenBMB | open | Estimated reference | 33.5% |
The top of this snapshot is led by Sakana Fugu-Ultra at 93.2%; third place is 3.6 points behind. The top-10 spread is 59.7 points.
About LiveCodeBench v6
LiveCodeBench v6 is a named release slice used in provider comparison tables. Keeping it separate prevents v6 results from being mixed into older or rolling LiveCodeBench windows. Results stay tied to the exact model variant and evaluation system. Multiple systems for the same model use the best published score on this page; overall Lumina scoring uses the median of ranking-eligible rows.
Open benchmark source ↗FAQ
LiveCodeBench v6 is a named release slice used in provider comparison tables. Keeping it separate prevents v6 results from being mixed into older or rolling LiveCodeBench windows.
Sakana Fugu-Ultra by Sakana AI currently leads with 93.2%.
10 models in the LuminaBench cohort have a qualifying score on this benchmark.
Related
