Benchmarks

Artificial Analysis Intelligence Index

General intelligence: knowledge, reasoning, math and coding rolled into one number

What it measures

Artificial Analysis is an independent evaluator. It runs a dozen public evaluations itself under one harness (knowledge Q&A, scientific reasoning, competition math, coding, terminal tasks) and blends the results into a single 0–100 index.

Higher means stronger overall. It answers “roughly where does this model sit?”

quickly, but it cannot tell you about any single skill. Each reasoning-effort setting (low / high / max) of a model is scored as its own entry.

Reading the scores

a relative 0–100 index, only meaningful among models scored in the same period; its composition shifts as evaluations are updated.

SourceArtificial Analysis

Model performance

Checked 20 results

Artificial Analysis Intelligence Index — Model performance. Select a column heading to sort.
Claude Fable 5.1 (max)Anthropic53
Claude Fable 5.1 (xhigh)Anthropic53
GPT-6 Astra (max)OpenAI53
GPT-6 Astra (xhigh)OpenAI53
Claude Fable 5.1 (high)Anthropic51
GPT-6 Astra (high)OpenAI51
Claude Opus 5 (max)Anthropic51
Claude Fable 5Anthropic50
GPT-6 Astra (medium)OpenAI50
Claude Opus 5 (xhigh)Anthropic50
Claude Fable 5.1 (medium)Anthropic49
Claude Opus 5 (high)Anthropic48
Muse Spark 1.3 (max)Meta48
GPT-5.6 Sol (max)OpenAI47
Claude Fable 5.1 (low)Anthropic47
GPT-6 Astra (low)OpenAI46
GPT-6 Astra (Non-reasoning)OpenAI45
Muse Spark 1.3 (xhigh)Meta45
Claude Opus 5 (medium)Anthropic45
GLM-5.3 (max)Z AI45

⌘KSearch the knowledge base

RECENTLY ADDED

Loading the knowledge index…