LiveCodeBench: scores, leaders and methodology
LiveCodeBench leaderboard data from the index. Contamination-free coding problems collected from recent contests. The top score is 85.1 and the average is 61.0.
What the data shows
Claude Sonnet 4.5 Reasoning leads at 85.1. The gap to the average is 24.1 points, and the gap to the 10th-ranked model is 1.5 points.
The median score is 64.7. Models above that line are in the upper half of the index for this benchmark.
How to read the ranking
- Use the score to shortlist, not to declare a winner.
- Check price and context for the top candidates.
- Read the benchmark methodology before quoting a score.
- Pair this benchmark with a task-specific evaluation.
Top 20 on LiveCodeBench
| # | Model | Provider | Score | Input / 1M | Context |
|---|---|---|---|---|---|
| 1 | Claude Sonnet 4.5 Reasoning | Anthropic | 85.1 | $19.48 | 400,000 |
| 2 | Grok 4 Reasoning | xAI | 85.1 | $19.67 | 2,000,000 |
| 3 | Claude Opus 4.5 | Anthropic | 85.0 | $19.42 | 400,000 |
| 4 | GPT-5.2 Max | OpenAI | 84.8 | $21.39 | 1,000,000 |
| 5 | Grok 4 Turbo | xAI | 84.7 | $19.21 | 1,000,000 |
| 6 | GPT-5.1 Turbo | OpenAI | 84.6 | $19.19 | 2,000,000 |
| 7 | GPT-5.2 Pro | OpenAI | 84.6 | $21.32 | 2,000,000 |
| 8 | GPT-5.2 Reasoning | OpenAI | 84.6 | $20.98 | 1,000,000 |
| 9 | Claude Opus 4.5 Pro | Anthropic | 84.1 | $20.98 | 200,000 |
| 10 | Gemini 3 Pro Max | Google DeepMind | 83.6 | $20.57 | 2,000,000 |
| 11 | GPT-5.2 Vision | OpenAI | 83.4 | $19.03 | 1,000,000 |
| 12 | o4 Reasoning | OpenAI | 83.4 | $19.69 | 1,000,000 |
| 13 | Gemini 3 Pro | Google DeepMind | 83.3 | $18.24 | 2,000,000 |
| 14 | Gemini 3 Pro Pro | Google DeepMind | 83.1 | $19.80 | 200,000 |
| 15 | Claude Sonnet 4.5 Max | Anthropic | 82.8 | $19.90 | 1,000,000 |
| 16 | Qwen3-Max Max | Alibaba Cloud | 82.8 | $18.53 | 400,000 |
| 17 | Gemini 3 Pro Turbo | Google DeepMind | 82.7 | $19.62 | 400,000 |
| 18 | o3 Reasoning | OpenAI | 82.7 | $18.07 | 200,000 |
| 19 | Claude Sonnet 4.5 | Anthropic | 82.5 | $17.84 | 400,000 |
| 20 | GPT-5.2 Turbo | OpenAI | 82.5 | $20.73 | 1,000,000 |
Methodology
This story uses the AI & LLM Index dataset: 600 models, 40 providers and 12 benchmarks. Prices are listed per million tokens. Scores are modelled estimates on a 0–100 scale and are not official benchmark results. Figures are rounded for readability.
Related
FAQ
What does the LiveCodeBench: scores, leaders and methodology data show?
LiveCodeBench leaderboard data from the index. Contamination-free coding problems collected from recent contests. The top score is 85.1 and the average is 61.0.
Are the numbers official?
No. Prices and scores are compiled for this demo index. Benchmark values are modelled estimates, not official results. Verify figures with the provider or benchmark owner before making decisions.
How should I use this ranking?
Use it to form a shortlist. Check the specific model pages, compare price and context, and test the shortlist on your own examples before committing.