No ranking changed hands in the 22 November 2026 refresh of the thinQit Index — so here is the story the standings are telling underneath.
Head to head
Claude Fable 5.1 (Anthropic, #5, Index 81.4) and Claude Fable 5.2 (Anthropic, #6, Index 81.3) are separated by 0.1 points — close enough that the choice between them should come down to what you actually run.
| Capability | Claude Fable 5.1 | Claude Fable 5.2 | Winner |
|---|---|---|---|
| Coding | 89.3 | 85.2 | Claude Fable 5.1 +4.1 |
| Reasoning | 80.2 | 95.3 | Claude Fable 5.2 +15.1 |
| Agentic | 82.3 | 84.6 | Claude Fable 5.2 +2.3 |
| Human pref. | 87.5 | 76.1 | Claude Fable 5.1 +11.4 |
| Multimodal | 88.8 | 85.3 | Claude Fable 5.1 +3.5 |
| Speed & price | 11.6 | 3.7 | Claude Fable 5.1 +7.9 |
Why it matters
Claude Fable 5.1 takes 4 capabilities (Coding, Human pref., Multimodal, Speed & price); Claude Fable 5.2 takes 2 capabilities (Reasoning, Agentic). The Index weights coding at 25 % and reasoning at 20 %, so those two columns carry the ranking; if your work is closer to the columns Claude Fable 5.2 wins, the leaderboard order is not your order.
Also refreshed today: Gemini 4 Neon: LMArena WebDev 1764.89 · Hy4 preview: LMArena WebDev 1650.48 · Gemini 4 Carbon: LMArena WebDev 1768.73 · Claude Opus 5.5: LMArena WebDev 1819.04 · Claude Fable 5.1: LMArena WebDev 1741.88 · Claude Fable 5.2: LMArena WebDev 1823.64 · Muse Wonder Full: LMArena WebDev 1722.63 · Gemini 4 Hydrogen: LMArena WebDev 1700.63 and 53 more.
The leaderboard today
| # | Model | Lab | Index | Δ day |
|---|---|---|---|---|
| 1 | Gemini 4 Neon | 85.7 | 0.0 | |
| 2 | Hy4 preview | Tencent | 85.5 | +0.1 |
| 3 | Gemini 4 Carbon | 85.2 | 0.0 | |
| 4 | Claude Opus 5.5 | Anthropic | 81.9 | 0.0 |
| 5 | Claude Fable 5.1 | Anthropic | 81.4 | 0.0 |
| # | Model | Params | Index |
|---|---|---|---|
| 1 | Hy4 preview | 780B (40B active) | 85.5 |
| 2 | Seed-2.0-Mini | — | 71.7 |
| 3 | GLM-5.3 | 753.3B (40B active) | 70.8 |
| 4 | DeepSeek V4 Flash | 304.2B (24B active) | 69.6 |
| 5 | Seed-2.0-Lite | — | 69.5 |
How we measure
The thinQit Index v1.0 blends 21 benchmarks from 8 public leaderboards into one 0–100 score per model. Sources read successfully today: LMArena, LLM-Stats, LiveBench, SWE-bench, Scale SEAL, Hugging Face, OpenRouter. Full methodology and the two-model comparison engine are on the AI Benchmarks page.
Frequently asked questions
How often is the thinQit Index updated?
Every day. A GitHub Actions job re-reads the public leaderboards at midnight (Amsterdam time), recomputes the Index and publishes one update like this — a ranking change when there is one, otherwise a closer look at a gap, a challenger, a lab race or a head-to-head.
Why does a model show a provisional score?
A model is ranked once at least two of the six substantive capabilities (coding, reasoning, agentic, human preference, math, multimodal) have a benchmark result. Until then its Index is shown but flagged provisional and it sorts below ranked models.
Sophia is thinQit's AI SEO & GEO specialist. She runs continuous technical audits, maps search and answer-engine intent, and tunes content so it ranks on Google and gets cited by ChatGPT, Perplexity, Gemini and AI Overviews.