JumalAIta IntelligenceIndependent evaluation

Independent evaluation

Independent benchmarks
for AI models.

JumalAIta Intelligence evaluates AI models under fixed, published conditions. Every benchmark page carries its exact prompt, its protocol, the recorded evidence and its limitations.

Benchmarks
2
Models tested
15
Evaluation
Independent
Last updated
2026-11-10

01 / Benchmarks

Benchmarks

Benchmark 01 / Creative intelligence

Expert-rankedActive

Creative Reimplementation Benchmark

One reference. One prompt. A complete audiovisual work. Each model receives a video capture of a 2001 demoscene production and rebuilds it for the web in a single shot. The original authors of the demo rank the results.

Standings

  1. 1Claude Fable 5.2Anthropic
  2. 2Claude Opus 5.5Anthropic
  3. 3Claude Fable 5.1Anthropic

Ranked by the original authors of the reference demo (members of Jumalauta). A ranking only: no scores.

Benchmark 02 / Creative intelligence

Expert-rankedPreliminary

Creative Synthesis Benchmark

Research Jumalauta. Learn its demo engine. Create an original demo. Each model uses jml-engine-webgl to make a new Jumalauta demo. Experts rank the works on how well each one works as a new Jumalauta demo and how well it uses JML Engine.

Standings

    No results have been published yet.

    Evaluation by experts familiar with Jumalauta demos. A ranking only: no scores.

    In development

    More benchmarks in development.

    New benchmarks are announced in the changelog when they are published.

    02 / Principles

    How we evaluate

    1. 01

      Identical conditions

      Every model in a benchmark receives the same prompt, the same inputs and the same settings.

    2. 02

      Published verbatim

      The exact prompt is printed on the benchmark page, word for word, so anyone can rerun it.

    3. 03

      Evidence on the page

      Model outputs are recorded and published next to the standings, so readers can judge for themselves.

    4. 04

      Gaps stay visible

      What has not been published is labeled as such. Nothing is estimated, rounded up or filled in.

    03 / Changelog

    Latest updates

    1. Results

      Claude Fable 5.2 ranked first

      Claude Fable 5.2, run in Claude Code, was placed first by the original authors. Every other model moves down one place.

    2. Benchmark

      Creative Synthesis Benchmark introduced

      A new task: research Jumalauta and jml-engine-webgl, then create an original audiovisual demo for expert ranking. The prompt and protocol are published; no results have been published yet.

    3. Results

      Mistral Medium 3.5 ranked 12th

      Mistral Medium 3.5, run in Mistral Vibe CLI, was evaluated and placed last by the original authors.

    4. Results

      Claude Haiku 4.5 evaluated

      The original authors ranked Claude Haiku 4.5 sixth. Updated ranking: 1. Claude Opus 5.5, 2. Claude Fable 5.1, 3. GPT-6 Astra, 4. GPT-6 Sol, 5. Claude Opus 4.7, 6. Claude Haiku 4.5.