A new model joined the thinQit Index in the 20 November 2026 refresh, led by Gemini 4 Neon.
Who joined
| # | Model | Lab | Index |
|---|---|---|---|
| 1 | Gemini 4 Neon | 86.1 |
Why it matters
Gemini 4 Neon takes the top of the frontier board with a lead of 2.3 Index points over #2 Hy4 preview. Its strongest capability is Human pref. at 93.2; 5 of 7 capabilities are covered.
Also changed today
- Hy4 preview slipped from #1 to #2 on the frontier leaderboard, passed by Gemini 4 Carbon (Index 83.8 → 83.8).
- Gemini 4 Carbon slipped from #2 to #3 on the frontier leaderboard, passed by Claude Opus 5.5 (Index 83.2 → 83.1).
- Claude Opus 5.5 slipped from #3 to #5 on the frontier leaderboard, passed by Claude Fable 5.2 and Claude Fable 5.1 (Index 82.1 → 82.1).
- GPT-6.2 Astra moved up from #6 to #4 on the frontier leaderboard, passing Claude Fable 5.2 and Claude Fable 5.1 (Index 80.4 → 82.7).
- Claude Fable 5.2 slipped from #4 to #6 on the frontier leaderboard, passed by Claude Fable 5.1 and GPT-6.2 Astra (Index 81.5 → 81.5).
- Claude Fable 5.1 slipped from #5 to #7 on the frontier leaderboard, passed by GPT-6.2 Astra and Muse Wonder Full (Index 81.4 → 81.4).
Also refreshed today: Gemini 4 Neon: LMArena Text 1529.68 · Gemini 4 Neon: LMArena WebDev 1763.9 · Gemini 4 Neon: LiveBench Agentic 82.6% · Gemini 4 Neon: LiveBench Reasoning 89.6% · Gemini 4 Neon: LiveBench Coding 89.7% · Gemini 4 Neon: Blended price 6 · Hy4 preview: LMArena WebDev 1631.47 · Gemini 4 Carbon: LMArena Text 1538.24 and 118 more.
The leaderboard today
| # | Model | Lab | Index | Δ day |
|---|---|---|---|---|
| 1 | Gemini 4 Neon | 86.1 | — | |
| 2 | Hy4 preview | Tencent | 83.8 | 0.0 |
| 3 | Gemini 4 Carbon | 83.1 | −0.1 | |
| 4 | GPT-6.2 Astra | OpenAI | 82.7 | +2.3 |
| 5 | Claude Opus 5.5 | Anthropic | 82.1 | 0.0 |
| # | Model | Params | Index |
|---|---|---|---|
| 1 | Hy4 preview | 780B (40B active) | 83.8 |
| 2 | Seed-2.0-Mini | — | 71.7 |
| 3 | GLM-5.3 | 753.3B (40B active) | 70.9 |
| 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.
