State of the leaderboard: week 46, 9–15 November 2026
Week in progress: data as of , from 5 daily updates · refreshed every six hours
The week in 6 sentences
- In week 46 of 2026 (9–15 November 2026), Claude Fable 5.2 (Anthropic) took the #1 spot from Claude Opus 5.5 on the BenchLeader Index at 72.7, 3.2 points ahead of Gemini 4 Carbon (69.5), as of 13 Nov 2026.
- 1 model entered the top 10 (Claude Fable 5.2 at #1), pushing out Claude Sonnet 5.5; 8 other models changed places.
- 7 new models arrived; the highest-ranked is Claude Fable 5.2 (Anthropic) at #1 with an index of 72.7.
- The biggest riser was Claude Fable 5.2, up 3.3 points to 72.7; the biggest faller DeepSeek V4 Pro, down 3.1 to 56.0.
- 9 models cut a list price and 3 models raised one; the largest cut was Qwen2.5 7B Instruct's output price, 75% lower.
- 343 new benchmark results were published, the most on MirrorCode (30).
Top 10 as of 13 Nov 2026
The BenchLeader Index, one row per model at its best reasoning effort, against the table at the end of the previous week. Left the top 10: Claude Sonnet 5.5.
| # | Model | Provider | Index | Last week | Move |
|---|---|---|---|---|---|
| 1 | Claude Fable 5.2 | Anthropic | 72.7 | – | new |
| 2 | Gemini 4 Carbon | 69.5 | 2 | – | |
| 3 | Claude Opus 5.5 | Anthropic | 68.4 | 1 | ↓2 |
| 4 | GPT-6 Astra | OpenAI | 67.1 | 3 | ↓1 |
| 5 | Claude Fable 5.1 | Anthropic | 66.6 | 4 | ↓1 |
| 6 | GPT-6.1 Sol | OpenAI | 66.6 | 5 | ↓1 |
| 7 | Claude Fable 5 | Anthropic | 66.3 | 6 | ↓1 |
| 8 | Gemini 4 Argon | 65.3 | 7 | ↓1 | |
| 9 | GPT-5.6 Sol | OpenAI | 65.3 | 8 | ↓1 |
| 10 | Claude Opus 5 | Anthropic | 64.9 | 9 | ↓1 |
New models and their first results
- Agnes 2.5 Pro Theta (Sapiens AI): #170 on the index at 50.8
- Claude Fable 5.2 (Anthropic): #1 on the index at 72.7. First results: 93.9% on GPQA Diamond; 95.1% on FrontierMath Tiers 1–3; 100.0% on FrontierMath Tier 4; 99.7% on OTIS Mock AIME.
- Muse Ember (Meta): #152 on the index at 52.1. First results: 18.3% on MLCR; 1608 on Arena WebDev; 72.7% on SimpleBench; 64.3% on APEX-Agents.
- Claude Mythos 5.2 (Anthropic): not enough independent results for an index yet
- U2 Flash (Unisound): not enough independent results for an index yet
- ERNIE-Light-DT-1.0 (Baidu): not enough independent results for an index yet
- Gemini 4 Hydrogen (Google): #15 on the index at 63.4. First results: 1490 on Arena Text; 1514 on Arena Hard Prompts; 1518 on Arena Coding; 1704 on Arena WebDev.
Biggest movers on the index
| Model | Provider | 9 Nov 2026 | 13 Nov 2026 | Change |
|---|---|---|---|---|
| Claude Fable 5.2 | Anthropic | 69.4 | 72.7 | +3.3 |
| DeepSeek V4 Pro | DeepSeek | 59.1 | 56.0 | -3.1 |
| Nemotron 3 Ultra | NVIDIA | 44.2 | 42.1 | -2.1 |
| Hy4 | Tencent | 55.3 | 56.7 | +1.4 |
| Kimi K2 Thinking | Moonshot AI | 53.3 | 52.0 | -1.3 |
| MiniMax-M2.1 | MiniMax | 55.0 | 53.8 | -1.2 |
| MiniMax-M2 | MiniMax | 52.6 | 51.5 | -1.1 |
| Gemini 4 Carbon | 68.5 | 69.5 | +1.0 | |
| GPT-6.1 Terra | OpenAI | 65.0 | 64.1 | -0.9 |
| Kimi K3.1 Pro | Moonshot AI | 60.4 | 61.3 | +0.9 |
Price changes
Cheaper (9)
- Qwen2.5 7B Instruct output price moved from $0.20/M to $0.050/M (75% down).
- MiMo-V2.5-Pro (no reasoning) output price moved from $1.85/M to $0.70/M (62% down).
- Nemotron 3 Nano Omni 30B A3B (thinking) input price moved from $0.50/M to $0.20/M (60% down).
- Gemini 4 Carbon (high) input price moved from $4.00/M to $2.00/M (50% down).
- Gemini 4 Argon (high) input price moved from $4.00/M to $2.00/M (50% down).
- Ripple-65B Preview input price moved from $0.20/M to $0.10/M (50% down).
- MiMo-V2.5 input price moved from $0.18/M to $0.11/M (39% down).
- Gemma 4 26B A4B (thinking) input price moved from $0.10/M to $0.090/M (10% down).
- Gemma 4 31B (no reasoning) input price moved from $0.16/M to $0.15/M (6% down).
More expensive (3)
- DeepSeek V4 Pro (high) input price moved from $0.66/M to $1.32/M (100% up).
- Gemma 4 26B A4B (no reasoning) input price moved from $0.12/M to $0.13/M (8% up).
- Llama 4 Maverick output price moved from $0.92/M to $0.97/M (5% up).
List prices per million tokens; each model's full price history is on its page and the LLM API pricing tracker.
Other weeks
- Week 45, 2–8 Nov 2026
- Week 44, 26 Oct–1 Nov 2026
- Week 43, 19–25 Oct 2026
- Week 42, 12–18 Oct 2026
- Week 41, 5–11 Oct 2026
- Week 40, 28 Sep–4 Oct 2026
- Week 39, 21–27 Sep 2026
- Week 38, 14–20 Sep 2026
- Week 37, 7–13 Sep 2026
Benchmark categories: AI agent benchmarks · AI coding benchmarks · Reasoning and math benchmarks · Multimodal AI benchmarks · New AI model releases
How this is built: the BenchLeader Index is recomputed with every update from independent benchmark publishers (see the methodology), and a maker's own launch numbers are never used. Past ranks are rebuilt from the daily index history of the models listed today.
Cite as: BenchLeader, “AI leaderboard this week, 9–15 Nov 2026: who is #1”, data as of 13 Nov 2026, https://www.benchleader.com/state.