Benchmark leaderboard
Objective scores from independent sources — separate from community sentiment. This is not a “best model” ranking; different sources use different methodologies, shown below.
Source: Artificial AnalysisSource: LMArena
Sort by:
Sorted by Intelligence (highest first)
| Select | Model & Provider | Intelligence | Coding | Reasoning | Speed | Price ($/M tok) | Last updated | Action |
|---|---|---|---|---|---|---|---|---|
 GPT-5 OpenAI | 73Top | 75 | 1432Top | 71 | $14 | Compare | ||
 Claude Opus Anthropic | 71 | 78 | 1421 | 62 | $21 | Compare | ||
GPT-5 Codex OpenAI | 70 | 79Top | — | 68 | $14 | Compare | ||
 Gemini Pro | 69 | 70 | 1409 | 95 | $5 | Compare | ||
Claude Sonnet Anthropic | 64 | 71 | 1398 | 89 | $6 | Compare | ||
Gemini Flash | 58 | 60 | — | 187Top | $0.9Top | Compare |