Menu

    Independent analysis of AI

    AI Models — State of AI Q3 2025
    State of AI (Highlights) — Q3 2025

    Independent analysis of AI

    Understand the AI landscape to choose the best model and provider for your use case

    Join our community on Discord! Access Report
    INTELLIGENCE
    Index Score • Higher is better
    SPEED
    Tokens/Sec • Higher is better
    PRICE
    USD/1M • Lower is better

    Common Comparisons

    How do the latest LLMs stack up?
    Compare Language Models →
    How do the latest OpenAI models compare?
    OpenAI Models Compared →
    Which model has the highest hallucination rate?
    AA-Omniscience Leaderboard →
    Who has the best Video Generation model?
    Video Arena →
    Which model is fastest with 100k token prompts?
    Long Context Latency →

    Intelligence

    29 of 395 models
    Artificial Analysis Intelligence Index
    v4.0 incorporates 10 evaluations (GDPval-AA, SciCode, etc.)
    Included Evaluations

    GDPval-AA, 𝜏²-Bench Telecom, Terminal-Bench Hard, SciCode, AA-LCR, AA-Omniscience, IFBench, Humanity's Last Exam, GPQA Diamond, CritPt

    Intelligence Evaluations

    Independent Analysis
    GDPval-AA
    Agentic Real-World Tasks
    Terminal-Bench Hard
    Agentic Coding
    AA-Omniscience
    Knowledge Accuracy
    GPQA Diamond
    Scientific Reasoning

    Openness

    Artificial Analysis Openness Index
    Availability & Transparency

    API Provider Performance

    Comparison of gpt-oss-120B (high)
    Output Speed vs. Price
    Provider Speed Leaderboard

    Explore

    Hardware Benchmarking
    Video Arena & Leaderboard
    Image Arena & Leaderboard
    Speech Arena & Leaderboard

    Subscribe