Best Models
Gemini 3 Pro Preview (high) is currently the highest-ranked model for LiveCodeBench in the available data.
Use the top-ranked model as a starting point, not an automatic decision.
Compare the top choices on price, speed, availability, and your own constraints.
Open the linked result page and methodology before quoting the ranking.
LiveCodeBench evaluates coding ability on recent programming problems, helping reduce benchmark memorization effects.
| Rank | Model | Provider | Value |
|---|---|---|---|
| #1 | Gemini 3 Pro Preview (high) | 91.7% | |
| #2 | Gemini 3 Flash Preview (Reasoning) | 90.8% | |
| #3 | DeepSeek V3.2 Speciale | DeepSeek | 89.6% |
| #4 | GPT-5.2 (medium) | OpenAI | 89.4% |
| #5 | GLM-4.7 (Reasoning) | Z AI | 89.4% |
This answer uses the exact same metric and live database ranking as the linked Easy Benchmarks leaderboard.
Source: Artificial Analysis LiveCodeBench. Updated: Jul 16, 2026.
| #6 | GPT-5.2 (xhigh) | OpenAI | 88.9% |
| #7 | gpt-oss-120b (high) | OpenAI | 87.8% |
| #8 | Claude Opus 4.5 (Reasoning) | Anthropic | 87.1% |
| #9 | GPT-5.1 (high) | OpenAI | 86.8% |
| #10 | MiMo-V2-Flash (Reasoning) | Xiaomi | 86.8% |
| #11 | DeepSeek V3.2 (Reasoning) | DeepSeek | 86.2% |
| #12 | o4-mini (high) | OpenAI | 85.9% |
| #13 | Gemini 3 Pro Preview (low) | 85.7% |
| #14 | Kimi K2 Thinking | Kimi | 85.3% |
| #15 | GPT-5.1 Codex (high) | OpenAI | 84.9% |
| #16 | GPT-5 (high) | OpenAI | 84.6% |
| #17 | GPT-5 Codex (high) | OpenAI | 84% |
| #18 | GPT-5 mini (high) | OpenAI | 83.8% |
| #19 | GPT-5.1 Codex mini (high) | OpenAI | 83.6% |
| #20 | Grok 4 Fast (Reasoning) | SpaceXAI | 83.2% |
