✳ the wire · analysis
Six AI leaderboards, four different leaders: Claude Opus 5.5 tops three, and Artificial Analysis's leader is not on a public API
An email when a new frontier model answers on a public API, about 15 minutes after members · free, no card

Six public AI leaderboards, read on November 5, put four different models first among the 45 that ArtificialWatch tracks. Claude Opus 5.5 leads three of them. The leader on Artificial Analysis, Gemini 4 Carbon, is not on a public API.
WHO LEADS WHERE
- LMArena: Claude Opus 5.5, ahead of GPT-6 Astra.
- SimpleBench: Claude Opus 5.5, ahead of Claude Fable 5.1.
- LLM Stats: Claude Opus 5.5 at 59.9, ahead of Gemini 4 Carbon at 58.8.
- LiveBench: Claude Fable 5.1, ahead of GPT-6 Astra by less than half a point.
- Vals Index: Claude Sonnet 5.5 at 70.85%, ahead of Gemini 4 Carbon at 70.45%.
- Artificial Analysis: Gemini 4 Carbon, ahead of Claude Opus 5.5.
Anthropic models hold five of the six top spots.
THE SAME MODEL, DIFFERENT RANKS
Only four of the 45 models are rated by all six boards: Claude Opus 5.5, Claude Sonnet 5.5, GPT-6 Astra and GPT-6.1 Sol. Claude Sonnet 5.5 is first on the Vals Index, third on LMArena and LLM Stats, fifth on LiveBench and Artificial Analysis, and 13th on SimpleBench. Claude Opus 5.5 is never lower than third.
THE LEADER YOU CANNOT BUY
Gemini 4 Carbon is first on Artificial Analysis and second on the Vals Index and LLM Stats. Google announced it on October 28 and opened it to cyber defenders and trusted partners; it has no public API, model id or price yet. On ArtificialWatch's bench index it is second, at 95 across the four boards that rate it.
THE INDEX
ArtificialWatch's bench index turns each board's score into a percentile and takes the median, so a model that ranks high everywhere beats one that wins a single board. Claude Opus 5.5 leads it at 98.1, rated by all six.
HOW WE COUNTED
From ArtificialWatch's bench index (artificialwatch.com/benchmarks, also at /data/bench-index.json), as each board was last read: November 5, 2026, and November 2 for Artificial Analysis. Ranks are among the 45 models ArtificialWatch tracks; a board can also rate models we don't. We checked the LLM Stats and Vals Index leaders on those boards' own pages.
Source: ArtificialWatch bench index (artificialwatch.com/benchmarks) ↗ · the bench index
Cite this page
Writing about this story? Copy a link or a citation.
<a href="https://artificialwatch.com/wire/six-ai-leaderboards-four-different-leaders">Six AI leaderboards, four different leaders: Claude Opus 5.5 tops three, and Artificial Analysis's leader is not on a public API</a>ArtificialWatch, "Six AI leaderboards, four different leaders: Claude Opus 5.5 tops three, and Artificial Analysis's leader is not on a public API", https://artificialwatch.com/wire/six-ai-leaderboards-four-different-leaders
