# LiveCodeBench — AI model rankings

Coding problems published so recently the AI can't have seen them in training — a contamination-free test of raw programming skill. Higher is better.

12 tracked models have a published LiveCodeBench score. Higher is better. Scores come from published lab reports and benchmark sources; recorded sources appear beside the scores.

## Ranking

| Rank | Model | Developer | Score | Source | Released |
| --- | --- | --- | --- | --- | --- |
| 1 | DeepSeek-V4-Pro | DeepSeek | 93.5% | [BenchLM](https://benchlm.ai), retrieved 2026-07-11 | Apr 24 2026 |
| 2 | Qwen3.8-Flash-Next | Qwen | 91.9% | Lab | Aug 26 2026 |
| 3 | DeepSeek-V4-Flash | DeepSeek | 91.6% | [BenchLM](https://benchlm.ai), retrieved 2026-07-11 | Apr 24 2026 |
| 3 | Qwen3.7-Max | Qwen | 91.6% | [BenchLM](https://benchlm.ai), retrieved 2026-08-10 | May 20 2026 |
| 5 | Qwen3.8-27B | Qwen | 90.3% | Lab | Aug 14 2026 |
| 6 | Kimi K2.6 | Moonshot AI | 89.6% | [BenchLM](https://benchlm.ai), retrieved 2026-07-11 | Apr 21 2026 |
| 6 | Qwen3.7-Plus | Qwen | 89.6% | [BenchLM](https://benchlm.ai), retrieved 2026-08-10 | Jun 1 2026 |
| 8 | Kimi K2.5 | Moonshot AI | 85% | [BenchLM](https://benchlm.ai), retrieved 2026-07-11 | Jan 27 2026 |
| 9 | GLM-4.7 | Z.ai | 84.9% | [BenchLM](https://benchlm.ai), retrieved 2026-08-24 | Dec 22 2025 |
| 10 | Nemotron Nano 2 | NVIDIA | 71.1% | Lab | Aug 18 2025 |
| 11 | Nemotron 3 Nano | NVIDIA | 68.3% | Lab | Dec 15 2025 |
| 12 | Llama Nemotron Ultra 253B | NVIDIA | 66.3% | Lab | Apr 8 2025 |


---

Canonical page: https://aireleasetracker.com/benchmark/livecodebench
Site index: https://aireleasetracker.com/llms.txt
Source: AI Release Tracker (https://aireleasetracker.com). Most benchmark scores come from lab launch material; gathered results identify the leaderboard that published them.