LiveCodeBench Pro: leaderboard
Competitive coding problems at extreme difficulty. 'Hard' here means 99.9% of competitive coders can't solve them. Shows how LLMs stack up against elite human programmers.
Metric: Rating (CF-style). Source: web.archive.org. Saturation forecast: Around April 2027. 58 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Deep Think | 3298 | #24 |
| 2 | Gemini 3.1 Pro (Preview) | 2887 | #58 |
| 3 | Gemini 3 Pro (Preview) | 2439 | #65 |
| 4 | GPT-5.2 (High) | 2393 | #110 (GPT-5.2) |
| 5 | Gemini 3 Flash (Preview) | 2316 | #82 |
| 6 | GPT-5.1 (2025-11-13) (High) | 2269 | #102 (GPT-5.1 (2025-11-13)) |
| 7 | GPT-5 (High) | 2176 | #98 (GPT-5) |
| 8 | O4 Mini (2025-04-16) (High) | 2092 | #178 (O4 Mini (2025-04-16)) |
| 9 | Gemini 2.5 Pro | 1769 | #151 |
| 10 | Gemini 2.5 Pro (Preview 03-25) | 1694 | #123 |
| 11 | O3 Mini (2025-01-31) | 1688 | #239 |
| 12 | Qwen 3 235B A22B 2507 (Thinking) | 1673 | #260 (Qwen 3 235B A22B 2507) |
| 13 | Qwen 3 Next 80B A3B (Thinking) | 1603 | #320 (Qwen 3 Next 80B A3B) |
| 14 | Claude Sonnet 4.5 (Thinking) | 1412 | #141 (Claude Sonnet 4.5) |
| 15 | Gemini 2.5 Flash (Preview 04-17) | 1334 | #234 |
No result here: #4 Claude Opus 5.5, #6 GPT-6 Astra, #7 Gemini 4 Argon.
Interactive version: theaggregate.ai/benchmark?slug=livecodebench-pro · How It Works · Data refreshed daily, snapshot 2026-10-21.