↑ Higher is betterscore535 models
Intelligence Benchmark
The Intelligence Benchmark measures overall model capability using a composite of evaluations across reasoning, coding, agentic tasks, instruction following, and multimodal understanding. Scores are derived from normalized evaluations on the Ethen Intelligence Index. Higher scores indicate stronger general-purpose capability.
Leading models
Top 8 ranked models by intelligence · score
1AClaude Fable 5 (with fallback)
602AClaude Opus 4.8 (max)
563AIGPT-5.5 (xhigh)
554AClaude Opus 4.7 (max)
545AClaude Sonnet 5 (max)
536AIGPT-5.5 (high)
537?GLM-5.2 (max)
518AIGPT-5.4 (xhigh)
51535 models with data15 without data
| Rank | Model | Provider | Type | Intelligence(score) |
|---|---|---|---|---|
| 1 | Claude Fable 5 (with fallback) | AAnthropic | Open weights | 60 |
| 2 | Claude Opus 4.8 (max) | AAnthropic | Open weights | 56 |
| 3 | GPT-5.5 (xhigh) | AIOpenAI | Open weights | 55 |
| 4 | Claude Opus 4.7 (max) | AAnthropic | Open weights | 54 |
| 5 | Claude Sonnet 5 (max) | AAnthropic | Open weights | 53 |
| 6 | GPT-5.5 (high) | AIOpenAI | Open weights | 53 |
| 7 | GLM-5.2 (max) | ?Z AI | Open weights | 51 |
| 8 | GPT-5.4 (xhigh) | AIOpenAI | Open weights | 51 |
| 9 | Gemini 3.5 Flash | GGoogle | Open weights | 50 |
| 10 | GPT-5.5 (medium) | AIOpenAI | Open weights | 50 |
| 11 | Claude Sonnet 4.6 (max) | AAnthropic | Open weights | 47 |
| 12 | Gemini 3.1 Pro Preview | GGoogle | Open weights | 46 |
| 13 | Qwen3.7 Max | QAlibaba | Open weights | 46 |
| 14 | Gemini 3.5 Flash (medium) | GGoogle | Open weights | 45 |
| 15 | Claude Opus 4.6 (max) | AAnthropic | Open weights | 44 |
| 16 | DeepSeek V4 Pro (max) | DDeepSeek | Open weights | 44 |
| 17 | GPT-5.3 Codex (xhigh) | AIOpenAI | Open weights | 44 |
| 18 | Kimi K2.6 | KKimi | Open weights | 44 |
| 19 | MiniMax-M3 | ?MiniMax | Open weights | 44 |
| 20 | Claude Opus 4.7 (Non-reasoning, high) | AAnthropic | Open weights | 43 |
| 21 | GPT-5.5 (low) | AIOpenAI | Open weights | 43 |
| 22 | Muse Spark | ?Meta | Open weights | 43 |
| 23 | Claude Sonnet 5 (Non-reasoning) | AAnthropic | Open weights | 42 |
| 24 | GPT-5.2 (xhigh) | AIOpenAI | Open weights | 42 |
| 25 | Kimi K2.7 Code | KKimi | Open weights | 42 |
| 26 | MiMo-V2.5-Pro | ?Xiaomi | Open weights | 42 |
| 27 | Claude Opus 4.5 | AAnthropic | Open weights | 41 |
| 28 | DeepSeek V4 Pro (high) | DDeepSeek | Open weights | 41 |
| 29 | Nex-N2-Pro | ?Nex AGI | Open weights | 41 |
| 30 | DeepSeek V4 Flash (max) | DDeepSeek | Open weights | 40 |
| 31 | Gemini 3 Pro Preview (high) | GGoogle | Open weights | 40 |
| 32 | GLM-5 | ?Z AI | Open weights | 40 |
| 33 | GLM-5.1 | ?Z AI | Open weights | 40 |
| 34 | GPT-5.2 Codex (xhigh) | AIOpenAI | Open weights | 40 |
| 35 | GPT-5.4 mini (xhigh) | AIOpenAI | Open weights | 40 |
| 36 | Grok Build 0.1 0616 | xxAI | Open weights | 40 |
| 37 | MiMo-V2.5 | ?Xiaomi | Open weights | 40 |
| 38 | MiMo-V2-Pro | ?Xiaomi | Open weights | 40 |
| 39 | Qwen3.6 Max Preview | QAlibaba | Open weights | 40 |
| 40 | Qwen3.6 Plus | QAlibaba | Open weights | 40 |
| 41 | GPT-5.4 (low) | AIOpenAI | Open weights | 39 |
| 42 | Qwen3.7 Plus | QAlibaba | Open weights | 39 |
| 43 | Claude Opus 4.6 (high) | AAnthropic | Open weights | 38 |
| 44 | Gemini 3 Flash | GGoogle | Open weights | 38 |
| 45 | GLM-5-Turbo | ?Z AI | Open weights | 38 |
| 46 | GPT-5.2 (medium) | AIOpenAI | Open weights | 38 |
| 47 | GPT-5.4 nano (xhigh) | AIOpenAI | Open weights | 38 |
| 48 | Grok 4.3 (high) | xxAI | Open weights | 38 |
| 49 | Kimi K2.5 | KKimi | Open weights | 38 |
| 50 | MiniMax-M2.7 | ?MiniMax | Open weights | 38 |
Page 1 of 11 · 550 results