Edition methodology 2026-10-19T00:00Z
Observed evaluations
This edition ranks models that have an admissible Arena evaluation, an admissible Artificial Analysis index, or both. A missing evaluation is not estimated and is not scored as 100. Two models at 100 may each be the best result of a different test. This score is not the corroborated v3 profile.
- Latest included observation
- Frozen calculation
- Publication
Calculation version : models-quality-observed
What the figures can tell us
Cumulative popularity does not measure recent growth. A benchmark depends on its tasks, date and model coverage. Scores are relative to the observed pool.
The team's opinions are signed and separate from the calculation. A Talk Show recommendation does not change a rank.
This archive keeps its historical version. The v3 method and its calibration are tracked separately and are not applied retroactively.
Preserved measurements and sources
The measurements below belong to this edition. A metric shown for context does not necessarily contribute to the score. Effective weights are shown only where they were preserved.
| Rank | Name and sources | Preserved measurements | Preserved effective weights |
|---|---|---|---|
| 1 | Claude Opus 5.5 (Max, Default Fallback) |
| Not recorded in this archive |
| 2 | Claude Sonnet 5.5 (Max, Default Fallback) |
| Not recorded in this archive |
| 3 | Claude Fable 5.1 (Max, Default Fallback) |
| Not recorded in this archive |
| 4 | Qwen3.8 Max |
| Not recorded in this archive |
| 5 | Gemini 4 Argon (High) |
| Not recorded in this archive |
| 6 | Muse Wonder (Max) |
| Not recorded in this archive |
| 7 | GPT-6 Astra (Max) |
| Not recorded in this archive |
| 8 | GPT-6.1 Sol (Max) |
| Not recorded in this archive |
| 9 | Claude Opus 5.5 (High) |
| Not recorded in this archive |
| 10 | Claude Fable 5 (Max, Opus 4.8 Fallback) |
| Not recorded in this archive |
| 11 | Muse Spark 1.3 (Max) |
| Not recorded in this archive |
| 12 | GPT-6 Sol (Max) |
| Not recorded in this archive |
| 13 | GPT-5.6 Sol (Max) |
| Not recorded in this archive |
| 14 | Claude Fable 5 (High) |
| Not recorded in this archive |
| 15 | Grok 4.7 (Xhigh) |
| Not recorded in this archive |
| 16 | MiMo-V2.6-Pro |
| Not recorded in this archive |
| 17 | GLM-5.3 (Max) |
| Not recorded in this archive |
| 18 | Claude Opus 4.6 (High) |
| Not recorded in this archive |
| 19 | Claude Haiku 5.5 (Max) |
| Not recorded in this archive |
| 20 | GPT-5.6 Terra (Max) |
| Not recorded in this archive |
Historical configuration
The parameters below were frozen at calculation time. They may contain planned dimensions absent from some rows. They do not replace the preserved measurements and effective weights.
Read the JSON configuration
{
"size": 20,
"basis": "models-quality-observed",
"cohort": 418,
"paired": 5,
"minimumCandidates": 20
}Observation, calculation and publication
An observation dates a measurement at a source. Calculation freezes the score using available observations. Publication makes the edition accessible. These dates can differ; a new collection does not automatically republish a ranking.
Scheduled calculation slots at 03:17 and 15:17 UTC do not guarantee a new edition: coverage, baselines and validation can delay it. The displayed date belongs to the edition actually published.