25 new models joined the thinQit Index in the 30 October 2026 refresh, led by Gemini 4 Carbon.
Who joined
| # | Model | Lab | Index |
|---|---|---|---|
| 1 | Gemini 4 Carbon | 86.6 | |
| 4 | Muse Wonder Full | Meta | 79.4 |
| 5 | Gemini 4 Argon | 79.2 | |
| 7 | GPT-6.1 Sol | OpenAI | 78.3 |
| 10 | Claude Sonnet 5.5 | Anthropic | 76.8 |
| 14 | GPT-6.1 Terra | OpenAI | 73.4 |
| 20 | Muse Wonder | Meta | 70.7 |
| 23 | Nemotron 3.5 Super | NVIDIA | 49.2 |
| 28 | Qwen3.7 Plus | Alibaba · Qwen | 67.4 |
| 31 | Kimi K3.1 Pro | Moonshot AI | 65.8 |
| 32 | GPT-6 Sol | OpenAI | 65.4 |
| 33 | Beam 1.0 | Reflection AI | 65.1 |
13 more models joined below these.
Why it matters
Gemini 4 Carbon takes the top of the frontier board with a lead of 2.8 Index points over #2 Claude Opus 5.5. Its strongest capability is Coding at 100.0; 4 of 7 capabilities are covered. 1 of the 25 newcomers is provisional, with fewer than two capabilities covered so far. Models enter the Index when the public leaderboards rank them, so a launch can show up a few days after the announcement.
Also changed today
- K2 Horizon lost 5.6 Index points (60.9 → 55.3) after new results: SWE-bench Verified new at 70.6%; AIME 2026 new at 48.5%; Output speed 969.17 → 928.97.
- Kimi K3 lost 2.4 Index points (65.6 → 63.2) after new results: AA Intelligence Index 43.6 → 17.4.
- Kimi K3.1 Pro gained 1.2 Index points (64.6 → 65.8) after new results: GPQA Diamond new at 94.2%; Humanity's Last Exam 52.5 → 51.1%; Output speed 32.42 → 32.65.
- GPT-5.6 Terra lost 1.0 Index points (67.7 → 66.7) after new results: Output speed 114.41 → 0.
- Mistral Large 4 gained 1.0 Index points (47.3 → 48.3) after new results: Output speed 106.95 → 169.06; Blended price 2.07 → 1.05; AA Intelligence Index 25.3 → 26.
- Qwen3.8 2.4T-A95B moved up from #8 to #7 on the local leaderboard, passing GLM-5.3-Flash (Index 67.9 → 67.9).
Also refreshed today: Gemini 4 Carbon: Output speed 87.74 · Claude Opus 5.5: Output speed 96.45 · Claude Fable 5.1: Output speed 67.85 · Gemini 4 Argon: Output speed 79.75 · Hy4 preview: Blended price 1.13 · GPT-6.1 Sol: Output speed 83.54 · Claude Sonnet 5.5: Output speed 134.42 · GPT-6.1 Terra: Output speed 92.27 and 51 more.
The leaderboard today
| # | Model | Lab | Index | Δ day |
|---|---|---|---|---|
| 1 | Gemini 4 Carbon | 86.6 | 0.0 | |
| 2 | Claude Opus 5.5 | Anthropic | 83.8 | 0.0 |
| 3 | Claude Fable 5.1 | Anthropic | 82.8 | 0.0 |
| 4 | Muse Wonder Full | Meta | 79.4 | 0.0 |
| 5 | Gemini 4 Argon | 79.2 | 0.0 |
| # | Model | Params | Index |
|---|---|---|---|
| 1 | Hy4 preview | 780B (40B active) | 78.9 |
| 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) | 69.8 |
| 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: Artificial Analysis, 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.
