Kaggle Open Benchmarks - LiveCodeBench V6: leaderboard
Pass rate on LiveCodeBench release V6, contest programming problems collected over time from LeetCode, AtCoder and Codeforces and judged by hidden tests. Kaggle's standardized re-run on its own harness under the open-benchmarks organization, not the original authors' leaderboard; Kaggle's fraction is shown as a percentage (0-100, higher is better). Earlier cumulative releases V1-V5 are subsets of V6 and are not published separately.
Metric: Pass Rate (%). Source: www.kaggle.com. Saturation forecast: Estimated already saturated. 44 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.5 Flash | 92.99 | #66 |
| 2 | Gemini 3 Pro (Preview) | 91.48 | #71 |
| 3 | Gemma 4 26B A4B | 90.7 | #360 |
| 4 | GPT-5 | 89.56 | #105 |
| 5 | Grok 4 | 86.39 | #189 |
| 6 | O3 | 85.52 | #132 |
| 7 | Gemma 4 31B | 84.84 | #212 |
| 8 | O4 Mini | 82.57 | #185 |
| 9 | Gemini 2.5 Pro (Preview 06-05) | 80.08 | #118 |
| 10 | GPT-OSS-120B | 78.64 | #346 |
| 11 | DeepSeek R1 0528 | 76.82 | #232 |
| 12 | O1 | 76.63 | #190 |
| 13 | O3 Mini | 76.44 | #282 |
| 14 | Claude Opus 4.8 | 74.23 | #39 |
| 15 | Gemini 2.5 Flash (Preview 05-20) | 70.69 | #229 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=kaggle-open-benchmarks-livecodebench-v6 · How It Works · Data refreshed daily, snapshot 2026-11-05.