Changes

11 releases · RSS

11 Nov 2026

2 new models

1 new model

6 new models, 2 index moves

New: Claude Opus 5.5 Parallel AdvancedClaude Opus 5.5 Parallel FastERNIE 5.5 Flash PreviewERNIE Light DT 1.0Mi:dm K 3.0 ProMotif 3
Moves: Gemini 4 Carbon80.7 → 82.7Qwen3.8 Max (0902)62.2 → 63.4

21 new models, 1 newly ranked, 1 left the ranking, 9 index moves

New: Claude Fable 5.285.5Inkling 2 Small66.3Muse Ember64.4Ripple-65B Preview52.4Celeris-1.1 DecisionClaude Mythos 5.2Doubao Seed 2.0 Lite (Apr 2026)Doubao Seed Evolving (Nov 2026)Drex MaxDrex Max v1.1Gemini 3 Pro (Refinement)Gemini Omni 1.1 FlashGemini Omni 1.2 FlashGLiDE v1.2GPT-5.2 (Refinement)GPT-6 CyberGrok 4 (Refinement)LongCat-2.5Nex N2.6PrincipledMachines Cortex 0.1+1 more
Ranked: Hy471.8
Left: Mistral Small 4 (Reasoning)28.1
Moves: Kimi K3.1 Pro77.7 → 75.8K-EXAONE 2.050.4 → 48.8Gemini 4 Carbon82.3 → 80.7Apodex 1.160.5 → 59.0Solar Pro 453.7 → 52.2Apodex 1.1 Mini57.1 → 55.8Granite 4.2 30B42.9 → 41.6Granite 4.2 8B38.6 → 37.6Granite 4.2 3B34.2 → 33.2

8 new models

New: Crouton AlphaDoubao Seed 1.6Doubao Seed 2.0 Lite 260428Doubao Seed 2.1 ProDoubao Seed 2.1 TurboDoubao Seed EvolvingDoubao-Seed-2.0 ProHy4

1 new model

2 new models

2 new models

10 index moves

Moves: Nemotron 3 Ultra52.9 → 51.2Mistral Large 459.3 → 60.9Claude Opus 5.580.2 → 78.7Claude Fable 5.177.8 → 76.3Beam 1.0 501B-A23B59.5 → 61.0Qwen3.8-27B61.2 → 59.9GPT-6 Astra79.8 → 78.6MiMo-V2.6-Pro69.4 → 70.6Kimi K3.1 Pro76.6 → 77.7Gemini 3.8 Flash69.6 → 68.6

27 new models, 5 newly ranked, 7 left the ranking, 10 index moves

New: Gemini 4 Carbon82.6GPT-6.1 Terra77.7Kimi K3.1 Pro76.6Qwen3.5-4B44.7LFM3-1.7B41.7d1-2Bd1-Omni-2BDeepSeek-V4-Pro (High, 20260813)DessaGPT-5.6 Sol KeenableGPT-6.1 Terra ProLFM3-1.7B-BaseLFM3-Audio-2BLFM3-Omni-2BLFM3-VL-2BNemotron 3.5 Bolt E4B BF16Nemotron Super VL-3.5 Nvfp4NVIDIA-Nemotron-3.5-Super-VL-120B-A12B (NVFP4)Particle-1Qwen3.8 Max Internal+7 more
Ranked: Beam 1.058.8Mercury 352.8Claude 4.1 Opus Thinking47.4Agents-A1-4B47.1Mistral Small 4 (Reasoning)28.1
Left: Hy356.2Quasar 438B52.5Ling 3.0 Flash VL51.2Ultravox v0.6 Llama 3.3 70B30.7Qwen3-Omni-30B-A3B-Thinking30.5Qwen3-Omni-30B-A3B-Instruct26.8LFM2.5-VL-1.6B-Extract19.2
Moves: Qwen3.5 397B (Reasoning)45.6 → 55.6Nemotron 3 Nano 30B39.5 → 33.2MiniMax M3.1 Flash72.7 → 70.3Llama 3.1 70B27.2 → 25.1Gemini 4 Argon81.0 → 79.5GPT-5.3 Codex64.0 → 62.6Llama 3.1 405B28.7 → 27.3LFM2.5-24B-A2B41.2 → 39.9Claude Opus 5.581.5 → 80.2Claude Fable 576.0 → 74.7

26 new models, 2 newly ranked, 2 index moves