AI leaderboard changes November 2026, so far
As of Nov 21, 2026, November 2026 has 302 confirmed crown changes: 42 on score slices (overall, open weights, size and similar) and 260 on best-per-dollar slices. The month also logged 146 newly announced models tracked, 18 tracked models with a first independent result and 69 source snapshots from 8 sources.1
Net changes by category
| Category | Slice | Changes | Holder on the 1st | Holder at month end | Days held | Margin |
|---|---|---|---|---|---|---|
| Coding | overall | 1 | Claude Opus 5.5 & Claude Sonnet 5.5 | Claude Opus 5.5 & Claude Sonnet 5.5 & Kimi K3.1 Pro | 9 | 0.00 |
| Coding | ≤32B | 1 | no holder | Mistral Large 4 | 11 | 0.00 |
| Coding | open weights | 2 | GLM 5.3 & GLM 5.3 Flash | GLM 5.3 & DeepSeek V4.1 Flash 1017 & Hy4 +1 | 6 | 0.00 |
| Coding | best per dollar | 32 | Claude Sonnet 5.5 & GLM 5.3 Flash & DeepSeek V4.1 Flash +2 | Claude Sonnet 5.5 & DeepSeek V4.1 Flash & DeepSeek V4.1 Flash 1017 +3 | 0 | 0.00 |
| Coding › Agentic | overall | 1 | Claude Opus 5.5 & Claude Sonnet 5.5 | Claude Opus 5.5 & Claude Sonnet 5.5 & Kimi K3.1 Pro | 9 | 0.00 |
| Coding › Agentic | ≤32B | 1 | no holder | Mistral Large 4 | 11 | 0.00 |
| Coding › Agentic | open weights | 2 | GLM 5.3 & GLM 5.3 Flash | GLM 5.3 & DeepSeek V4.1 Flash 1017 & Hy4 +1 | 6 | 0.00 |
| Coding › Agentic | best per dollar | 24 | Claude Opus 5.5 & Claude Sonnet 5.5 & GLM 5.3 +1 | Claude Haiku 5.5 & Claude Opus 5.5 & Claude Sonnet 5.5 +4 | 0 | 0.00 |
| Coding › Web dev | overall | 1 | Claude Opus 5.5 | Claude Fable 5.2 & Claude Opus 5.5 | 10 | 0.00 |
| Coding › Web dev | open weights | 2 | DeepSeek V4.1 Flash & Kimi K3 | Kimi K3 & DeepSeek V4.1 Flash & Hy4 +1 | 1 | 0.00 |
| Coding › Web dev | best per dollar | 32 | Claude Sonnet 5.5 & DeepSeek V4.1 Flash & GLM 5.3 Flash +2 | Claude Sonnet 5.5 & DeepSeek V4.1 Flash & Gemini 4 Carbon +3 | 1 | 0.00 |
| Cybersecurity | best per dollar | 20 | Claude Haiku 5.5 & Gemini 3.8 Flash & GLM 5.3 +4 | Claude Haiku 5.5 & Gemini 3.8 Flash & GLM 5.3 +5 | 1 | 0.00 |
| Decision models | overall | 4 | Fastino GLiDE 1.2 & Sieve 31B v3 & KnowLine 31B Gen1 +1 | GLiDE 27B & karpo-decision-31B & kudhab (Gemma 4 31B) +1 | 0 | 0.00 |
| Decision models | calibration | 2 | no holder | Copernicus-57B | 4 | 0.00 |
| Decision models | same-GPU latency | 1 | no holder | Cerno 27B v1.1 | 4 | 0.00 |
| Decision models | ≤32B | 3 | Fastino GLiDE 1.2 & Sieve 31B v3 & KnowLine 31B Gen1 +1 | GLiDE 27B & karpo-decision-31B & kudhab (Gemma 4 31B) +1 | 0 | 0.00 |
| Decision models | open weights | 4 | Fastino GLiDE 1.2 & Sieve 31B v3 & KnowLine 31B Gen1 +1 | GLiDE 27B & karpo-decision-31B & kudhab (Gemma 4 31B) +1 | 0 | 0.00 |
| Factuality | overall | 2 | Gemini 4 Argon & Claude Opus 5.5 & Muse Wonder | Gemini 4 Carbon & Gemini 4 Neon | 2 | 0.00 |
| Factuality | best per dollar | 32 | Mimo V2.6 Pro & gpt-oss-120b & GLM 5.3 Flash +3 | DeepSeek V4.1 Flash & Gemini 4 Carbon & Gemini 4 Hydrogen +3 | 0 | 0.00 |
| Finance | best per dollar | 9 | GLM 5.3 Flash & GPT-5.6 Luna & Muse Spark 1.1 +1 | GLM 5.3 Flash & GPT-5.6 Luna & Muse Spark 1.1 +1 | 32 | 0.00 |
| General LLM | overall | 1 | Gemini 4 Argon & Muse Wonder & Claude Opus 5.5 | Gemini 4 Carbon & Gemini 4 Argon & Muse Wonder | 13 | 0.00 |
| General LLM | best per dollar | 37 | Gemini 3.7 Flash & Gemini 4 Argon & Hy3 +4 | Gemini 3.7 Flash & Gemini 3.8 Flash & Gemini 4 Argon +3 | 11 | 0.00 |
| Image edit | open weights | 1 | Qwen Image Edit | MiniMax H3-Image | 4 | 0.00 |
| Image generation | open weights | 2 | Ideogram 4.0 Quality | Ideogram 4.0 & MiniMax H3-Image | 4 | 0.00 |
| Long context | overall | 2 | Gemini 4 Argon & Claude Opus 5.5 | Gemini 4 Argon & Gemini 4 Carbon & Gemini 4 Neon | 2 | 0.00 |
| Long context | open weights | 1 | Kimi K3 & Mimo V2.6 Pro | Kimi K3 & Mimo V2.6 Pro & GLM 5.3 | 12 | 0.00 |
| Long context | best per dollar | 32 | Gemini 3.7 Flash & Gemini 3.8 Flash & Gemini 4 Argon +2 | Gemini 3.7 Flash & Gemini 3.8 Flash & Gemini 4 Argon +4 | 1 | 0.00 |
| Reasoning | overall | 3 | Gemini 4 Argon & Claude Opus 5.5 & Muse Wonder | Gemini 4 Carbon & Gemini 4 Argon & Gemini 4 Neon | 2 | 0.00 |
| Reasoning | ≤32B | 1 | Gemma 4 31B & Qwen3.8 27B & Gemma 4 26B A4b | Gemma 4 31B & Gemma 4 26B A4b | 15 | 0.00 |
| Reasoning | open weights | 1 | Kimi K3 & Mimo V2.6 Pro | Kimi K3 & GLM 5.3 & Mimo V2.6 Pro | 13 | 0.00 |
| Reasoning | best per dollar | 22 | Gemini 3.8 Flash & Mimo V2.6 Pro & Gemini 3.7 Flash +2 | Gemini 3.7 Flash & Gemini 3.8 Flash & Gemini 4 Argon +3 | 1 | 0.00 |
| Video generation | overall | 1 | Sora 2 | Gemini Omni 1.2 Flash & Gemini Omni 1.1 Flash | 13 | 0.00 |
| Video generation | open weights | 1 | Wan V2.2 A14b | Kandinsky 5.0 Video Pro | 15 | 0.00 |
| Video generation | best per dollar | 1 | no holder | Gemini Omni 1.2 Flash | 15 | 0.00 |
| Vision | overall | 1 | no holder | no holder | — | — |
| Vision | best per dollar | 19 | GPT-6 Luna & Muse Spark 1.3 & GLM 5.3 Flash +3 | DeepSeek V4.1 Flash & Gemini 3.8 Flash & Gemini 4 Carbon +3 | 1 | 0.00 |
Confirmed crown changes
- Nov 21, 2026. Decision models: GLiDE 27B and karpo-decision-31B and kudhab (Gemma 4 31B) and Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout) are co-SOTA. GLiDE 27B, karpo-decision-31B, kudhab (Gemma 4 31B), Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout) · Decision models
- Nov 21, 2026. Decision models: GLiDE 27B and karpo-decision-31B and kudhab (Gemma 4 31B) and Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout) are co-SOTA. GLiDE 27B, karpo-decision-31B, kudhab (Gemma 4 31B), Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout) · Decision models
- Nov 21, 2026. Decision models: GLiDE 27B and karpo-decision-31B and kudhab (Gemma 4 31B) and Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout) are co-SOTA. GLiDE 27B, karpo-decision-31B, kudhab (Gemma 4 31B), Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout) · Decision models
- Nov 20, 2026. Coding › Web dev: Kimi K3 and DeepSeek V4.1 Flash and Hy4 and Inkling 2 Small are co-SOTA. Kimi K3, DeepSeek V4.1 Flash, Hy4, Inkling 2 Small · Coding › Web dev
- Nov 19, 2026. Reasoning: Gemini 4 Carbon and Gemini 4 Argon and Gemini 4 Neon are co-SOTA. Gemini 4 Carbon, Gemini 4 Argon, Gemini 4 Neon · Reasoning
- Nov 19, 2026. Long context: Gemini 4 Argon and Gemini 4 Carbon and Gemini 4 Neon are co-SOTA. Gemini 4 Argon, Gemini 4 Carbon, Gemini 4 Neon · Long context
- Nov 19, 2026. Factuality: Gemini 4 Carbon and Gemini 4 Neon are co-SOTA. Gemini 4 Carbon, Gemini 4 Neon · Factuality
- Nov 17, 2026. Image generation: Ideogram 4.0 and MiniMax H3-Image are co-SOTA. Ideogram 4.0, MiniMax H3-Image · Image generation
- Nov 17, 2026. Image edit: MiniMax H3-Image takes the crown. MiniMax H3-Image · Image edit
- Nov 17, 2026. Decision models: Cerno 27B v1.1 takes the crown. Cerno 27B v1.1 · Decision models
- Nov 17, 2026. Decision models: Copernicus-57B takes the crown. Copernicus-57B · Decision models
- Nov 15, 2026. Coding › Agentic: GLM 5.3 and DeepSeek V4.1 Flash 1017 and Hy4 and GLM 5.3 Flash are co-SOTA. GLM 5.3, DeepSeek V4.1 Flash 1017, Hy4, GLM 5.3 Flash · Coding › Agentic
- Nov 15, 2026. Coding: GLM 5.3 and DeepSeek V4.1 Flash 1017 and Hy4 and GLM 5.3 Flash are co-SOTA. GLM 5.3, DeepSeek V4.1 Flash 1017, Hy4, GLM 5.3 Flash · Coding
- Nov 12, 2026. Decision models: GLiDE 27B and karpo-decision-31B and Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout) are co-SOTA. GLiDE 27B, karpo-decision-31B, Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout) · Decision models
- Nov 12, 2026. Coding › Agentic: Claude Opus 5.5 and Claude Sonnet 5.5 and Kimi K3.1 Pro are co-SOTA. Claude Opus 5.5, Claude Sonnet 5.5, Kimi K3.1 Pro · Coding › Agentic
- Nov 12, 2026. Coding: Claude Opus 5.5 and Claude Sonnet 5.5 and Kimi K3.1 Pro are co-SOTA. Claude Opus 5.5, Claude Sonnet 5.5, Kimi K3.1 Pro · Coding
- Nov 12, 2026. Decision models: GLiDE 27B and karpo-decision-31B and Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout) are co-SOTA. GLiDE 27B, karpo-decision-31B, Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout) · Decision models
- Nov 12, 2026. Coding › Agentic: GLM 5.3 and DeepSeek V4.1 Flash 1017 and GLM 5.3 Flash are co-SOTA. GLM 5.3, DeepSeek V4.1 Flash 1017, GLM 5.3 Flash · Coding › Agentic
- Nov 12, 2026. Coding: GLM 5.3 and DeepSeek V4.1 Flash 1017 and GLM 5.3 Flash are co-SOTA. GLM 5.3, DeepSeek V4.1 Flash 1017, GLM 5.3 Flash · Coding
- Nov 11, 2026. Reasoning: Gemini 4 Carbon and Gemini 4 Argon are co-SOTA. Gemini 4 Carbon, Gemini 4 Argon · Reasoning
- Nov 11, 2026. Decision models: REAX 1.3 and GLiDE 27B and karpo-decision-31B and Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout) are co-SOTA. REAX 1.3, GLiDE 27B, karpo-decision-31B, Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout) · Decision models
- Nov 11, 2026. Coding › Web dev: Claude Fable 5.2 and Claude Opus 5.5 are co-SOTA. Claude Fable 5.2, Claude Opus 5.5 · Coding › Web dev
- Nov 11, 2026. Decision models: REAX 1.3 and GLiDE 27B and karpo-decision-31B and Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout) are co-SOTA. REAX 1.3, GLiDE 27B, karpo-decision-31B, Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout) · Decision models
- Nov 11, 2026. Decision models: GLiDE 27B and karpo-decision-31B and Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout) are co-SOTA. GLiDE 27B, karpo-decision-31B, Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout) · Decision models
- Nov 11, 2026. Decision models: GLiDE 9B takes the crown. GLiDE 9B · Decision models
- Nov 10, 2026. Coding › Agentic: Mistral Large 4 takes the crown. Mistral Large 4 · Coding › Agentic
- Nov 10, 2026. Coding: Mistral Large 4 takes the crown. Mistral Large 4 · Coding
- Nov 9, 2026. Long context: Kimi K3 and Mimo V2.6 Pro and GLM 5.3 are co-SOTA. Kimi K3, Mimo V2.6 Pro, GLM 5.3 · Long context
- Nov 9, 2026. Coding › Web dev: DeepSeek V4.1 Flash and Kimi K3 and Inkling 2 Small are co-SOTA. DeepSeek V4.1 Flash, Kimi K3, Inkling 2 Small · Coding › Web dev
- Nov 8, 2026. Video generation: Gemini Omni 1.2 Flash and Gemini Omni 1.1 Flash are co-SOTA. Gemini Omni 1.2 Flash, Gemini Omni 1.1 Flash · Video generation
- Nov 8, 2026. Reasoning: Gemini 4 Carbon and Gemini 4 Argon and Muse Wonder are co-SOTA. Gemini 4 Carbon, Gemini 4 Argon, Muse Wonder · Reasoning
- Nov 8, 2026. Long context: Gemini 4 Argon and Gemini 4 Carbon are co-SOTA. Gemini 4 Argon, Gemini 4 Carbon · Long context
- Nov 8, 2026. General LLM: Gemini 4 Carbon and Gemini 4 Argon and Muse Wonder are co-SOTA. Gemini 4 Carbon, Gemini 4 Argon, Muse Wonder · General LLM
- Nov 8, 2026. Factuality: Gemini 4 Carbon takes the crown. Gemini 4 Carbon · Factuality
- Nov 8, 2026. Reasoning: Kimi K3 and GLM 5.3 and Mimo V2.6 Pro are co-SOTA. Kimi K3, GLM 5.3, Mimo V2.6 Pro · Reasoning
- Nov 6, 2026. Decision models: REAX 1.3 and GLiDE 27B and Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout) are co-SOTA. REAX 1.3, GLiDE 27B, Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout) · Decision models
- Nov 6, 2026. Video generation: Kandinsky 5.0 Video Pro takes the crown. Kandinsky 5.0 Video Pro · Video generation
- Nov 6, 2026. Decision models: REAX 1.3 and GLiDE 27B and Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout) are co-SOTA. REAX 1.3, GLiDE 27B, Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout) · Decision models
- Nov 6, 2026. Reasoning: Gemma 4 31B and Gemma 4 26B A4b are co-SOTA. Gemma 4 31B, Gemma 4 26B A4b · Reasoning
- Nov 6, 2026. Decision models: GLiDE 27B and Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout) are co-SOTA. GLiDE 27B, Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout) · Decision models
- Nov 5, 2026. Vision: Muse Spark 1.3 and Claude Sonnet 5.5 and GPT-6.1 Sol and Gemini 4 Argon and Muse Wonder are co-SOTA. Muse Spark 1.3, Claude Sonnet 5.5, GPT-6.1 Sol, Gemini 4 Argon, Muse Wonder · Vision
- Nov 4, 2026. Image generation: Ideogram 4.0 takes the crown. Ideogram 4.0 · Image generation
Price-slice reshuffles
Best-per-dollar crowns move with price as well as score, so their co-SOTA lists reshuffle often. One line per category.
- General LLM (best per dollar): 37 reshuffles among Gemini 3.7 Flash, Gemini 3.8 Flash, Gemini 4 Argon, Gemini 4 Carbon, GLM 5.3, GLM 5.3 Flash and 9 more.
- Coding (best per dollar): 32 reshuffles among Claude Sonnet 5.5, DeepSeek V4.1 Flash, DeepSeek V4.1 Flash 1017, GLM 5.3 Flash, GPT-6.1 Terra, Kimi K3.1 Pro and 5 more.
- Coding › Web dev (best per dollar): 32 reshuffles among Claude Sonnet 5.5, DeepSeek V4.1 Flash, Gemini 4 Carbon, GLM 5.3 Flash, GPT-6.1 Terra, Qwen4 and 8 more.
- Factuality (best per dollar): 32 reshuffles among DeepSeek V4.1 Flash, Gemini 4 Carbon, Gemini 4 Hydrogen, GLM 5.3, GLM 5.3 Flash, Mimo V2.6 Pro and 9 more.
- Long context (best per dollar): 32 reshuffles among Gemini 3.7 Flash, Gemini 3.8 Flash, Gemini 4 Argon, Gemini 4 Carbon, GLM 5.3, GLM 5.3 Flash and 6 more.
- Coding › Agentic (best per dollar): 24 reshuffles among Claude Haiku 5.5, Claude Opus 5.5, Claude Sonnet 5.5, GLM 5.3, GLM 5.3 Flash, GPT-6.1 Terra and 4 more.
- Reasoning (best per dollar): 22 reshuffles among Gemini 3.7 Flash, Gemini 3.8 Flash, Gemini 4 Argon, Gemini 4 Carbon, GLM 5.3, GLM 5.3 Flash and 9 more.
- Cybersecurity (best per dollar): 20 reshuffles among Claude Haiku 5.5, Gemini 3.8 Flash, GLM 5.3, GPT-6.1 Sol, GPT-6 Astra, Kimi K3.1 Pro and 3 more.
- Vision (best per dollar): 19 reshuffles among Gemini 3.8 Flash, Gemini 4 Carbon, GLM 5.3, GPT-6 Luna, Muse Spark 1.3, DeepSeek V4.1 Flash and 3 more.
- Finance (best per dollar): 9 reshuffles among GLM 5.3 Flash, GPT-5.6 Luna, Muse Spark 1.1, Muse Spark 1.2, DeepSeek V4 Flash 0731, GPT-5 Nano.
- Video generation (best per dollar): 1 reshuffle among Gemini Omni 1.2 Flash.
Newly tracked and first benchmarked
- Nov 21, 2026: Voxa3D V1.5 announced and tracked, pending benchmarks
- Nov 20, 2026: Aion 1.0 Instruct announced and tracked, pending benchmarks
- Nov 20, 2026: bison-1 announced and tracked, pending benchmarks
- Nov 20, 2026: groundhog-1 announced and tracked, pending benchmarks
- Nov 20, 2026: Inkling 2 Mini announced and tracked, pending benchmarks
- Nov 19, 2026: Gemini 4 Neon got its first independent result
- Nov 19, 2026: Olala mini 7A1B announced and tracked, pending benchmarks
- Nov 19, 2026: Tell-1 announced and tracked, pending benchmarks
- Nov 18, 2026: AssemblyAI-STT-3.7 announced and tracked, pending benchmarks
- Nov 18, 2026: AssemblyAI-STT-3.7-Realtime announced and tracked, pending benchmarks
- Nov 18, 2026: AssemblyAI-STT-3.7-Realtime-Agent announced and tracked, pending benchmarks
- Nov 18, 2026: Gemini 4 Neon announced and tracked, pending benchmarks
- Nov 18, 2026: GPT-6.2 Sol got its first independent result
- Nov 18, 2026: HeyGen Avatar 6 announced and tracked, pending benchmarks
- Nov 18, 2026: HeyGen Avatar 6 Turbo announced and tracked, pending benchmarks
- Nov 18, 2026: Hy4 Flash got its first independent result
- Nov 18, 2026: Kling Image 4.0 announced and tracked, pending benchmarks
- Nov 18, 2026: SID-2 announced and tracked, pending benchmarks
- Nov 18, 2026: SID-2 Swarm announced and tracked, pending benchmarks
- Nov 17, 2026: GPT-6.2 Astra announced and tracked, pending benchmarks
- Nov 17, 2026: GPT-6.2 Astra got its first independent result
- Nov 17, 2026: GPT-6.2 Sol announced and tracked, pending benchmarks
- Nov 17, 2026: MAI Code 2 announced and tracked, pending benchmarks
- Nov 17, 2026: MAI Code 2 Flash announced and tracked, pending benchmarks
- Nov 17, 2026: MAI Image 2.7 Flash announced and tracked, pending benchmarks
- Nov 17, 2026: MAI Realtime 1 announced and tracked, pending benchmarks
- Nov 17, 2026: Microsoft-Decision-1.1 announced and tracked, pending benchmarks
- Nov 17, 2026: Qwen4-Max got its first independent result
- Nov 16, 2026: d1.2 announced and tracked, pending benchmarks
- Nov 16, 2026: d1.2-ultra announced and tracked, pending benchmarks
- Nov 16, 2026: Hy4 Flash announced and tracked, pending benchmarks
- Nov 16, 2026: Qwen4-Max announced and tracked, pending benchmarks
- Nov 16, 2026: Sage S2.1 Flash announced and tracked, pending benchmarks
- Nov 15, 2026: SOV-Transcribe-1 announced and tracked, pending benchmarks
- Nov 14, 2026: K-EXAONE 2.5 announced and tracked, pending benchmarks
- Nov 13, 2026: Gemini 4 Hydrogen got its first independent result
- Nov 13, 2026: Handoff² announced and tracked, pending benchmarks
- Nov 13, 2026: MiniMax H3-Image got its first independent result
- Nov 13, 2026: Raon-1 nano announced and tracked, pending benchmarks
- Nov 13, 2026: Taan v1-preview announced and tracked, pending benchmarks
- Nov 12, 2026: Cartwheel Control announced and tracked, pending benchmarks
- Nov 12, 2026: ci-proto announced and tracked, pending benchmarks
- Nov 12, 2026: Edge0-120B announced and tracked, pending benchmarks
- Nov 12, 2026: Edge0-145B announced and tracked, pending benchmarks
- Nov 12, 2026: Gemini 4 Hydrogen announced and tracked, pending benchmarks
- Nov 12, 2026: Mango 4 announced and tracked, pending benchmarks
- Nov 12, 2026: persona-2.0-live announced and tracked, pending benchmarks
- Nov 12, 2026: Tonality announced and tracked, pending benchmarks
- Nov 12, 2026: WeatherNext 3 announced and tracked, pending benchmarks
- Nov 11, 2026: AlphaGenome 2 announced and tracked, pending benchmarks
- Nov 11, 2026: BotPsy-0.8B announced and tracked, pending benchmarks
- Nov 11, 2026: ERNIE 5.5 Flash Preview announced and tracked, pending benchmarks
- Nov 11, 2026: Gemini Decision 1 announced and tracked, pending benchmarks
- Nov 11, 2026: Praxis-1 announced and tracked, pending benchmarks
- Nov 11, 2026: Qwen4-Max-Preview announced and tracked, pending benchmarks
- Nov 11, 2026: Simons announced and tracked, pending benchmarks
- Nov 10, 2026: Alpamayo 2 Edge announced and tracked, pending benchmarks
- Nov 10, 2026: Fish Audio Live 3 announced and tracked, pending benchmarks
- Nov 10, 2026: MiniMax H3-Image announced and tracked, pending benchmarks
- Nov 10, 2026: Muse Ember got its first independent result
- Nov 10, 2026: Nemotron 3.5 Super VL got its first independent result
- Nov 10, 2026: NovaAtom-Lite announced and tracked, pending benchmarks
- Nov 10, 2026: PAN 2 announced and tracked, pending benchmarks
- Nov 10, 2026: Scribe v3 announced and tracked, pending benchmarks
- Nov 10, 2026: Scribe v3 Live announced and tracked, pending benchmarks
- Nov 10, 2026: ZONOS3-Mini announced and tracked, pending benchmarks
- Nov 9, 2026: Claude Fable 5.2 announced and tracked, pending benchmarks
- Nov 9, 2026: Claude Fable 5.2 got its first independent result
- Nov 9, 2026: GR00T 2 announced and tracked, pending benchmarks
- Nov 9, 2026: LingBot-Touch announced and tracked, pending benchmarks
- Nov 9, 2026: Muse Ember announced and tracked, pending benchmarks
- Nov 9, 2026: OrcaC2 1.0 announced and tracked, pending benchmarks
- Nov 9, 2026: Seedream 5.5 announced and tracked, pending benchmarks
- Nov 9, 2026: Sophon-Realtime v1 announced and tracked, pending benchmarks
- Nov 8, 2026: LearningUnit 2.0 (L1-30B-A5B) announced and tracked, pending benchmarks
- Nov 8, 2026: MNN-Pocket-0.3B announced and tracked, pending benchmarks
- Nov 8, 2026: Morgana v1.2 announced and tracked, pending benchmarks
- Nov 8, 2026: Youtu-Parsing-Omni announced and tracked, pending benchmarks
- Nov 7, 2026: Alpie Seed announced and tracked, pending benchmarks
- Nov 7, 2026: Inkling 2 Small got its first independent result
- Nov 7, 2026: PrimeSim announced and tracked, pending benchmarks
- Nov 7, 2026: Qoduck announced and tracked, pending benchmarks
- Nov 7, 2026: Underdog Woof announced and tracked, pending benchmarks
- Nov 7, 2026: Zuse announced and tracked, pending benchmarks
- Nov 6, 2026: Arete-27B announced and tracked, pending benchmarks
- Nov 6, 2026: Gemini Omni 1.2 Flash got its first independent result
- Nov 6, 2026: HyperFlow H3 Flash announced and tracked, pending benchmarks
- Nov 6, 2026: Kanana-2-v-decide announced and tracked, pending benchmarks
- Nov 6, 2026: Manta announced and tracked, pending benchmarks
- Nov 6, 2026: Mirage announced and tracked, pending benchmarks
- Nov 6, 2026: PLaMo 3 Translate announced and tracked, pending benchmarks
- Nov 6, 2026: Quark announced and tracked, pending benchmarks
- Nov 6, 2026: REAX 1.3 got its first independent result
- Nov 6, 2026: S5 announced and tracked, pending benchmarks
- Nov 6, 2026: S5-nano announced and tracked, pending benchmarks
- Nov 6, 2026: XERJ Query-1 announced and tracked, pending benchmarks
- Nov 5, 2026: Accentus announced and tracked, pending benchmarks
- Nov 5, 2026: ERNIE-5.5-Flash-Base announced and tracked, pending benchmarks
- Nov 5, 2026: Gemini Omni 1.2 Flash announced and tracked, pending benchmarks
- Nov 5, 2026: GLiDE announced and tracked, pending benchmarks
- Nov 5, 2026: GPT-6 Cyber announced and tracked, pending benchmarks
- Nov 5, 2026: GPT-Image-2.6 Flare got its first independent result
- Nov 5, 2026: GPT-Image-2.6 Sunburst got its first independent result
- Nov 5, 2026: Higgs Realtime Duplex announced and tracked, pending benchmarks
- Nov 5, 2026: Hy4 got its first independent result
- Nov 5, 2026: Inkling 2 Small announced and tracked, pending benchmarks
- Nov 5, 2026: Ling-3.2-flash announced and tracked, pending benchmarks
- Nov 5, 2026: Orvena 4B announced and tracked, pending benchmarks
- Nov 5, 2026: THX-02 announced and tracked, pending benchmarks
- Nov 5, 2026: VINCO announced and tracked, pending benchmarks
- Nov 4, 2026: AIRoA-0 (A-0) announced and tracked, pending benchmarks
- Nov 4, 2026: GPT-Image-2.6 Flare announced and tracked, pending benchmarks
- Nov 4, 2026: GPT-Image-2.6 Sunburst announced and tracked, pending benchmarks
- Nov 4, 2026: GroundingPI announced and tracked, pending benchmarks
- Nov 4, 2026: Hy4 announced and tracked, pending benchmarks
- Nov 4, 2026: IOTA Decide 4B announced and tracked, pending benchmarks
- Nov 4, 2026: Kandinsky S2V 1.0 Lite announced and tracked, pending benchmarks
- Nov 4, 2026: Kandinsky S2V 1.0 Pro announced and tracked, pending benchmarks
- Nov 4, 2026: Kling 4.0 Flash announced and tracked, pending benchmarks
- Nov 4, 2026: Liber-0 announced and tracked, pending benchmarks
- Nov 4, 2026: LiquidAI d1-2B got its first independent result
- Nov 4, 2026: LiquidAI d1-Omni-2B got its first independent result
- Nov 4, 2026: Pangram Image 1.0 announced and tracked, pending benchmarks
- Nov 4, 2026: Pangram Image Flash 1.0 announced and tracked, pending benchmarks
- Nov 4, 2026: Pulse Alpha announced and tracked, pending benchmarks
- Nov 4, 2026: Pyramid-JiT 2B (pjit-21k) announced and tracked, pending benchmarks
- Nov 4, 2026: Ripple-65B Preview announced and tracked, pending benchmarks
- Nov 4, 2026: Seldon announced and tracked, pending benchmarks
- Nov 4, 2026: SenseNova 6.8 Flash announced and tracked, pending benchmarks
- Nov 4, 2026: SWE-2 Flash announced and tracked, pending benchmarks
- Nov 4, 2026: TwIL-8B announced and tracked, pending benchmarks
- Nov 4, 2026: Verne V0.5 announced and tracked, pending benchmarks
- Nov 4, 2026: Viggle Remake announced and tracked, pending benchmarks
- Nov 3, 2026: Crouton Alpha announced and tracked, pending benchmarks
- Nov 3, 2026: CUA-Master-27B announced and tracked, pending benchmarks
- Nov 3, 2026: Endeavor 1.1 announced and tracked, pending benchmarks
- Nov 3, 2026: HeyGen Voice Turbo announced and tracked, pending benchmarks
- Nov 3, 2026: hs-decide-1-flash announced and tracked, pending benchmarks
- Nov 3, 2026: hs-embed-1 announced and tracked, pending benchmarks
- Nov 3, 2026: hs-rerank-1 announced and tracked, pending benchmarks
- Nov 3, 2026: Moondream Parakeet Ultra announced and tracked, pending benchmarks
- Nov 3, 2026: P-H3-Image announced and tracked, pending benchmarks
- Nov 3, 2026: Palette announced and tracked, pending benchmarks
- Nov 3, 2026: REAX 1.3 announced and tracked, pending benchmarks
- Nov 3, 2026: Sage2 announced and tracked, pending benchmarks
- Nov 3, 2026: Sarvam Prism announced and tracked, pending benchmarks
- Nov 3, 2026: Smaug-Audio-Decisions-0.7B announced and tracked, pending benchmarks
- Nov 3, 2026: Surgeon announced and tracked, pending benchmarks
- Nov 3, 2026: TaoMate-HD Preview announced and tracked, pending benchmarks
- Nov 3, 2026: Wave Forcing 1.3B announced and tracked, pending benchmarks
- Nov 3, 2026: WorldSonus announced and tracked, pending benchmarks
- Nov 2, 2026: ChefNMR2 announced and tracked, pending benchmarks
- Nov 2, 2026: Drex for Finance announced and tracked, pending benchmarks
- Nov 2, 2026: Hume Audio Decision Model announced and tracked, pending benchmarks
- Nov 2, 2026: Hydrogen-1.0 announced and tracked, pending benchmarks
- Nov 2, 2026: LongCat-2.5 announced and tracked, pending benchmarks
- Nov 2, 2026: LTX-2.5 announced and tracked, pending benchmarks
- Nov 2, 2026: Nori-Causal announced and tracked, pending benchmarks
- Nov 2, 2026: pplx-embed-v2-context announced and tracked, pending benchmarks
- Nov 2, 2026: VisionPsy-Nano-460M announced and tracked, pending benchmarks
- Nov 1, 2026: Mellum2.1 announced and tracked, pending benchmarks
- Nov 1, 2026: Solar Mini 4 announced and tracked, pending benchmarks
- Nov 1, 2026: Utopai X announced and tracked, pending benchmarks
- Nov 1, 2026: WeVisCut announced and tracked, pending benchmarks
Models that entered the top 3
- Claude Fable 5 at #1 in Abstraction (ARC)
- Claude Fable 5.2 at #1 in Coding › Web dev
- Claude Fable 5.2 at #1 in Abstraction (ARC)
- Claude Fable 5.2 at #1 in Finance
- Claude Opus 5 at #1 in Healthcare
- Claude Opus 5.5 at #1 in Coding
- Claude Opus 5.5 at #1 in Coding › Agentic
- Claude Opus 5.5 at #1 in Coding › Web dev
- Claude Opus 5.5 at #1 in Abstraction (ARC)
- Claude Sonnet 5.5 at #1 in Coding
- Claude Sonnet 5.5 at #1 in Coding › Agentic
- Gemini 3.8 Flash at #1 in Abstraction (ARC)
- Gemini 4 Argon at #1 in General LLM
- Gemini 4 Argon at #1 in Reasoning
- Gemini 4 Argon at #1 in Long context
- Gemini 4 Argon at #1 in Abstraction (ARC)
- Gemini 4 Carbon at #1 in General LLM
- Gemini 4 Carbon at #1 in Reasoning
- Gemini 4 Carbon at #1 in Long context
- Gemini 4 Carbon at #1 in Factuality
- Gemini 4 Carbon at #1 in Abstraction (ARC)
- Gemini 4 Carbon at #1 in Legal
- Gemini 4 Neon at #1 in General LLM
- Gemini 4 Neon at #1 in Reasoning
- Gemini 4 Neon at #1 in Long context
- Gemini 4 Neon at #1 in Factuality
- Gemini Omni 1.1 Flash at #1 in Video generation
- Gemini Omni 1.2 Flash at #1 in Video generation
- GLiDE 27B at #1 in Decision models
- GPT-6 Astra at #1 in Science
- GPT-6 Astra at #1 in Abstraction (ARC)
- GPT-6 Astra at #1 in Cybersecurity
- GPT-6.1 Sol at #1 in Abstraction (ARC)
- GPT-6.2 Astra at #1 in Science
- GPT-6.2 Astra at #1 in Abstraction (ARC)
- Jcorners Ingot 8B R3 at #1 in Embeddings
- karpo-decision-31B at #1 in Decision models
- Kimi K3.1 Pro at #1 in Coding
- Kimi K3.1 Pro at #1 in Coding › Agentic
- kudhab (Gemma 4 31B) at #1 in Decision models
- Muse Spark 1.2 at #1 in Finance
- Muse Wonder at #1 in General LLM
- Muse Wonder at #1 in Reasoning
- Muse Wonder at #1 in Cybersecurity
- Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout) at #1 in Decision models
- REAX 1.3 at #1 in Decision models
- Zoom Scribe V2.1 Pro at #1 in Speech-to-text
- Bytedance Seed1 6 Embedding 1215 at #2 in Embeddings
- Claude Fable 5 at #2 in Long context
- Claude Fable 5.1 at #2 in Math
- Claude Fable 5.1 at #2 in Legal
- Claude Fable 5.2 at #2 in Science
- Claude Fable 5.2 at #2 in Coding
- Claude Opus 4.6 Thinking at #2 in General LLM
- Claude Opus 5 at #2 in Legal
- Claude Opus 5.5 at #2 in Factuality
- Claude Opus 5.5 at #2 in Finance
- Claude Sonnet 5.5 at #2 in Coding › Web dev
- Gemini 3.1 Pro Preview at #2 in Abstraction (ARC)
- Gemini 3.1 Pro Preview at #2 in Healthcare
- Gemini 4 Hydrogen at #2 in Science
- Gemini Omni Flash at #2 in Video generation
- GPT Image 2.5 Flare at #2 in Image edit
- GPT-6 Astra at #2 in Coding › Agentic
- GPT-6.1 Sol at #2 in Coding › Agentic
- GPT-6.1 Sol at #2 in Cybersecurity
- GPT-Image-2.6 Sunburst at #2 in Image generation
- Kingsoft LLM Qzhou Embedding at #2 in Embeddings
- Microsoft Azure Speech 09 2026 at #2 in Speech-to-text
- Microsoft Azure Speech 09 2026new at #2 in Speech-to-text
- Muse Spark 1.1 at #2 in Finance
- Muse Spark 1.3 at #2 in Legal
- Muse Wonder at #2 in Reasoning
- Quyet-1.0-Large (Chinh Nguyen, Gemma-4-31B decoder, option-letter logits) at #2 in Decision models
- Zoom Scribe V2 Pro at #2 in Speech-to-text
- Claude Fable 5.1 at #3 in Coding › Agentic
- Claude Fable 5.1 at #3 in Factuality
- Claude Fable 5.1 at #3 in Abstraction (ARC)
- Claude Opus 4.6 at #3 in Reasoning
- Claude Opus 4.6 at #3 in Long context
- Claude Opus 4.7 Thinking at #3 in General LLM
- Claude Opus 5 at #3 in Abstraction (ARC)
- Claude Opus 5.5 at #3 in General LLM
- Claude Opus 5.5 at #3 in Science
- Flux 3 Video at #3 in Video generation
- Gemini 3.1 Pro Preview at #3 in Math
- Gemini 4 Argon at #3 in Cybersecurity
- Gemini 4 Argon at #3 in Finance
- Gemini 4 Argon at #3 in Legal
- Gemini 4 Argon at #3 in Healthcare
- Gemini 4 Carbon at #3 in Science
- Gemini 4 Carbon at #3 in Coding › Web dev
- GPT Image 2.5 Flare at #3 in Image generation
- GPT-5.6 Sol at #3 in Abstraction (ARC)
- GPT-6 Astra at #3 in Coding
- GPT-6.2 Sol at #3 in Abstraction (ARC)
- GPT-Image-2.6 Sunburst at #3 in Image edit
- Mai Image 2.5 Preview at #3 in Image generation
- Mai Image 2.6 Preview at #3 in Image edit
- Musci Research Musci ASR Qwen 1019 HF at #3 in Speech-to-text
- Musci Research Musci ASR Qwen 1019 HF New at #3 in Speech-to-text
- Muse Spark at #3 in Finance
- Nvidia Llama Embed Nemotron 8B at #3 in Embeddings
- REAX 1.3 at #3 in Decision models
Benchmarks added or saturated
- LMArena Text (style-controlled, overall). Updated Nov 19, 2026.
- LMArena Text (style-controlled, coding). Updated Nov 19, 2026.
- LMArena Text (style-controlled, math). Updated Nov 19, 2026.
- LMArena Text (style-controlled, hard prompts). Updated Nov 19, 2026.
- LMArena Text (style-controlled, longer query). Updated Nov 19, 2026.
- LMArena Vision (style-controlled). Updated Nov 19, 2026.
- LMArena WebDev. Updated Nov 20, 2026.
- LMArena Text-to-Image. Updated Nov 17, 2026.
- LMArena Text-to-Video. Updated Nov 9, 2026.
- LMArena Image Edit. Updated Nov 17, 2026.
- LMArena Text factuality. Updated Nov 19, 2026.
- GPQA Diamond. Updated Nov 18, 2026.
- SimpleQA Verified. Updated Nov 18, 2026.
- Terminal-Bench 4.0. Updated Nov 18, 2026.
- SWE-bench Pro V2 HARD. Updated Nov 6, 2026.
- GDP.pdf. Updated Nov 18, 2026.
- MTEB Multilingual v2 (MMTEB). Updated Nov 21, 2026.
- MTEB English v2. Updated Nov 21, 2026.
- Open ASR Leaderboard (English short-form). Updated Nov 21, 2026.
- ARC-AGI. Updated Nov 17, 2026.
- Decision Index Full. Updated Nov 21, 2026.
- Decision Index ECE. Updated Nov 21, 2026.
- Decision Index Brier. Updated Nov 21, 2026.
- Decision Index latency. Updated Nov 21, 2026.
- Vals Index. Updated Nov 21, 2026.
- Vibe Code Bench. Updated Nov 21, 2026.
- Vibe Code Bench 1-100. Updated Nov 15, 2026.
- Code Migration. Updated Nov 21, 2026.
- MysteryMechanism. Updated Nov 21, 2026.
- CorpFin. Updated Nov 13, 2026.
- MortgageTax. Updated Nov 13, 2026.
- ProofBench. Updated Nov 21, 2026.
- CyberBench. Updated Nov 21, 2026.
- SRE Bench. Updated Nov 21, 2026.
- Finance Agent. Updated Nov 21, 2026.
- Excel Modeling Benchmark (EMB). Updated Nov 21, 2026.
- TaxEval. Updated Nov 13, 2026.
- Tax Agent Bench. Updated Nov 21, 2026.
- Legal Research Bench. Updated Nov 21, 2026.
- MedCode. Updated Nov 21, 2026.
- MedScribe. Updated Nov 21, 2026.
1 Crown change confirmed by two consecutive snapshots of price and score. A rules-version recompute, a seed, and a refresh with unchanged source fingerprints do not count as a snapshot. Primary sources disagree beyond the margin of error, so the crown is shared with low confidence. Crown change confirmed by two consecutive snapshots or two independent primary sources. First holder recorded for this category and slice.