Model A GPT-6.1 Terra (OpenAI) Kimi K3.1 Pro (Moonshot AI) Cohere Rerank 5 (Cohere) Mercury 3 (Inception) MiniMax M3.1 Flash (MiniMax) DeepSeek V4.1 Flash 1017 (DeepSeek) Glimpse (Elorian) Moondream 3.1 (Moondream) Muse Spark Open (Meta AI) Moondream 3 Preview (Moondream) Nemotron 3.5 Super VL (NVIDIA) ScaleDown: Summarize (ScaleDown) LFM2.5-24B-A2B (Liquid AI) Pareto 26.9 (Unbiased) Muse Wonder (Meta AI) Claude Haiku 5.5 (Anthropic) Mistral Large 4 (Mistral AI) Nano Banana 2.1 (Google DeepMind) Ling 3.1 Flash (inclusionAI) Gemini 4 Argon (Google DeepMind) GPT-6.1 Sol (OpenAI) Claude Sonnet 5.5 (Anthropic) Perceptron Mk1.5 (Perceptron) Ember-1 (Fireworks AI) Aion 3.5 (Aion Labs) Aion 3.5 Mini (Aion Labs) GLM-5.3-Prime (Zhipu AI) Solar Mini 4 (Upstage) Claude Opus 5.5 (Anthropic) GPT-6 Luna (OpenAI) GPT-6 Sol (OpenAI) MiMo-V2.6-Pro-UltraSpeed (Xiaomi) MiMo-V2.6-Pro (Xiaomi) Grok 4.7 (SpaceXAI (formerly xAI)) MiMo-V2.6-Flash (Xiaomi) Step 5 Preview (StepFun) Qwen3.8-Omni-Flash (Alibaba / Qwen) Ternary Bonsai 2 27B (PrismML) GLM-5.3-FlashX (Zhipu AI) Schematron V2 Small (Inference.net) Schematron V2 Turbo (Inference.net) Fugu Ultra v2 (Sakana AI) Sakana: Fugu Max (Sakana AI) DeepSeek-V4.1-Flash (DeepSeek) Nex-N2.5-mini (Nex-AGI) Nex-N2.5-Pro (Nex-AGI) Mercury 2.5 (Inception) Ling 3.0 Flash VL (inclusionAI) GPT-6 Astra (OpenAI) Meta: Muse Spark 1.3 Contributor (Meta AI) Gemini 3.8 Flash (Google DeepMind) Meta: Muse Spark 1.3 (Meta AI) Claude Fable 5.1 (Anthropic) Tencent Hy4 Preview (Tencent) GLM-5.3-Flash (Zhipu AI) Qwen3.8 Flash (Alibaba / Qwen) Granite 4.2 8B (IBM) DeepSeek V4 Flash Vision Exp (DeepSeek) Hy-MT2-30B-A3B (Tencent) Qwen3.8-27B (Alibaba / Qwen) GLM-5.3 (Zhipu AI) Gemini 3.7 Flash (Google DeepMind) DeepSeek V4 Pro 0813 (DeepSeek) Qwen3.8-2.4T-A95B (Alibaba / Qwen) Grok 4.6 (SpaceXAI (formerly xAI)) Solar Pro 4 (Upstage) Sakana Namazu (Sakana AI) Nemotron 3.5 Lightning (NVIDIA) Seed-2.0-Code (ByteDance) Seed 2.1 Turbo (ByteDance) Muse Glimmer (Meta AI) Meta Muse Glimmer 30B (Meta AI) Muse Spark 1.2 (Meta AI) LFM2.5-2.6B (Liquid AI) Qwen3.8-Max (Alibaba / Qwen) DeepSeek V4 Flash 0731 (DeepSeek) Inkling Small (Thinking Machines Lab) Qwen3.7 Flash (Alibaba / Qwen) Claude Opus 5 (Anthropic) Ling-3.0-flash (inclusionAI) Laguna S 2.1 (Poolside) LongCat 2.0 (Meituan) Kimi K3 (Moonshot AI) Inkling (Thinking Machines Lab) Muse Spark 1.1 (Meta AI) GPT-5.6 Luna Pro (OpenAI) GPT-5.6 Luna (OpenAI) GPT-5.6 Terra Pro (OpenAI) GPT-5.6 Terra (OpenAI) GPT-5.6 Sol (OpenAI) GPT-5.6 Sol Pro (OpenAI) Grok 4.5 (SpaceXAI (formerly xAI)) Aion-3.0-Mini (Aion Labs) Aion-3.0 (Aion Labs) Hy3 (Tencent) Laguna XS 2.1 (Poolside) Gemini 3.1 Flash Lite Image (Google DeepMind) Claude Sonnet 5 (Anthropic) Fugu Ultra (Sakana AI) Gemini 3.1 Flash Image (Google DeepMind) Nano Banana Pro (Gemini 3 Pro Image) (Google DeepMind) GLM-5.2 (Zhipu AI) Kimi K2.7 Code (Moonshot AI) Claude Fable 5 (Anthropic) Nemotron 3.5 Content Safety (NVIDIA) Nemotron 3 Ultra (NVIDIA) Qwen3.7 Plus (Alibaba / Qwen) MiniMax-M3 (MiniMax) Claude Opus 4.8 (Anthropic) Step-3.7-Flash (StepFun) Hy-MT2-7B (Tencent) Hy-MT2-1.8B (Tencent) Gemini 3.5 Flash (Google DeepMind) Qwen3.7 Max (Alibaba / Qwen) Command A+ (Cohere) Grok Build 0.1 (SpaceXAI (formerly xAI)) Gemini Omni Flash (Google DeepMind) Perceptron Mk1 (Perceptron) Ring-2.6-1T (inclusionAI) Gemini 3.1 Flash Lite (Google DeepMind) Laguna M.1 (Poolside) Mistral Medium 3.5 (Mistral AI) Qwen3.6 Flash (Alibaba / Qwen) GPT-5.5 Pro (OpenAI) GPT-5.5 (OpenAI) Qwen3.6 27B (Alibaba / Qwen) DeepSeek V4 Pro (DeepSeek) DeepSeek V4 Flash (DeepSeek) GPT-5.4 Image 2 (OpenAI) Qwen3.6 Max Preview (Alibaba / Qwen) Kimi K2.6 (Moonshot AI) Grok 4.3 (SpaceXAI (formerly xAI)) Claude Opus 4.7 (Anthropic) Qwen3.6 35B A3B (Alibaba / Qwen) Gemma 4 26B A4B (Google DeepMind) GLM-5V-Turbo (Zhipu AI) Qwen 3.6 Plus (Alibaba / Qwen) Gemma 4 31B (Google DeepMind) Trinity Large Thinking (Arcee AI) Grok 4.20 Multi-Agent (SpaceXAI (formerly xAI)) Reka Edge (Reka) MiMo-V2-Omni (Xiaomi) MiMo-V2-Pro (Xiaomi) GPT-5.4 Nano (OpenAI) GPT-5.4 mini (OpenAI) Mistral Small 4 (Mistral AI) MiniMax M2.7 (MiniMax) GLM 5 Turbo (Zhipu AI) Nemotron 3 Super (NVIDIA) GPT-5.4 (OpenAI) Qwen3.5-9B (Alibaba / Qwen) Qwen3.5-27B (Alibaba / Qwen) Qwen3.5-35B-A3B (Alibaba / Qwen) Qwen3.5-Flash (Alibaba / Qwen) Mercury 2 (Inception) Gemini 3.1 Pro (Google DeepMind) Grok 4.20 (SpaceXAI (formerly xAI)) GLM-5.1 (Zhipu AI) Claude Sonnet 4.6 (Anthropic) Qwen3.5 397B A17B (Alibaba / Qwen) Seed-2.0-Mini (ByteDance) Seed-2.0-Lite (ByteDance) Doubao Seed 2.0 (ByteDance) MiniMax-M2.5 (MiniMax) GLM 5 (Zhipu AI) GPT-5.3-Codex (OpenAI) Claude Opus 4.6 (Anthropic) Step-3.5-Flash (StepFun) Kimi K2.5 (Moonshot AI) GLM-4.7 (Zhipu AI) Gemini 3 Flash (Google DeepMind) GPT-5.2 (OpenAI) Claude Opus 4.5 (Anthropic) Kimi K2 Thinking (Moonshot AI) MiniMax-M2 (MiniMax) Nanbeige4.3-33B (Nanbeige) Raon-1 mini preview (Krafton) MOSS-Image (Openmoss) Ling-3.0-flash-Sante (inclusionAI) Sage-A1.5 (Shanghai AI Lab) OrcaCyber Nano 1.0 (OrcaRouter) Reflection Beam 1.0 (Reflection AI) NVIDIA Nemotron 3.5 Super VL 120B-A12B NVFP4 (NVIDIA) Beam (Reflection AI) Clef (Cloudflare) Space Bunny Alpha (Undisclosed (stealth models)) MiMo-V2.6-Pro-RL (Xiaomi) MiMo-V2.6-Flash-RL (Xiaomi) FrogNano-4B-2609 (Microsoft) AliceAI-Foundation-80B-A3B-Base (Yandex) Xing4.0-29B-A4B (China Telecom AI (TeleAI)) Union Alpha (Undisclosed (stealth models)) Iris-pro (Allspark) Atria Dawn (Shanghai AI Lab) Granite Time Series PatchTST-FM-r2 (IBM) Ling 3.0 Flash Fin (inclusionAI) Qwen3.8-Flash-Next (Alibaba / Qwen) IBM Granite 4.2 (30B) (IBM) Granite 4.2 (IBM) Ox Alpha (OpenRouter) MiniMax Music 3 (MiniMax) Shieldstral (Mistral AI) K-EXAONE 2.0 (750B-A37B) (LG AI Research) LFM2.5-Encoder (Liquid AI) Qwen-Image-3.0 (Alibaba / Qwen) Nemotron-3-Embed-1B-BF16 (NVIDIA) Nemotron 3 Embed 8B BF16 (NVIDIA) KAT-Coder-Air V2.5 (Kwaipilot) KAT-Coder-Pro V2.5 (Kwaipilot) Audex-30B-A3B (NVIDIA) NVIDIA Nemotron-Labs-3-Puzzle-75B-A9B-NVFP4 (NVIDIA) Leanstral 1.5 (Mistral AI) LFM2.5-230M (Liquid AI) DeepSeek-V4-Fable (DeepSeek) Qwen-AgentWorld-35B-A3B (Alibaba / Qwen) Unlimited-OCR (Baidu AI) North Mini Code (Cohere) FastContext-1.0-4B-SFT (Microsoft) DiffusionGemma (Google DeepMind) North Mini Code 1.0 (Cohere) Nex-N2-Pro (Nex-AGI) Gemma 4 12B Unified (Google DeepMind) Mellum2-12B-A2.5B-Thinking (JetBrains) Cosmos3-Nano (NVIDIA) Cosmos 3 Super (NVIDIA) Cosmos 3 Super Image2Video (NVIDIA) Cosmos3-Super-Text2Image (NVIDIA) Voxtral Small (Mistral AI) LocateAnything-3B (NVIDIA) Fara1.5-27B (Microsoft) Fara-7B (Microsoft) Granite Embedding 97M Multilingual R2 (IBM) Baidu Qianfan: CoBuddy (Baidu AI) Granite Speech 4.1 2B (IBM) Granite 4.1 30B (IBM) Granite 4.1 8B (IBM) Granite Embedding 311M Multilingual R2 (IBM) Laguna XS.2 (Poolside) Nemotron 3 Nano Omni 30B-A3B-Reasoning (NVIDIA) Owl Alpha (OpenRouter) Nemotron-3-Nano-Omni-30B-A3B (NVIDIA) Ling-2.6-1T (inclusionAI) Trinity Large Preview (Arcee AI) OpenAI Privacy Filter (OpenAI) Gemma 4 E2B VLA (Google DeepMind) Pareto Code Router (OpenRouter) Ling-2.6-flash (inclusionAI) LFM2.5-VL-450M (Liquid AI) HY-Embodied-0.5 MoT-2B (Tencent) Harrier (27B) (Microsoft) Amazon Nova 2 Sonic (Amazon Web Services) Gemma 4 E4B Instruction-Tuned (Google DeepMind) Gemma 4 31B Dense (Google DeepMind) Gemma 4 E2B (Google DeepMind) Harrier-OSS v1 270M (Microsoft) Qwen3.5-Omni (Alibaba / Qwen) Google Lyria 3 Clip Preview (Google DeepMind) KAT-Coder-Pro V2 (Kwaipilot) gpt-oss-puzzle-88B (NVIDIA) Qwen3.5-Max-Preview (Alibaba / Qwen) RakutenAI 3.0 (Rakuten) NVIDIA Nemotron-3-Nano-4B-GGUF (NVIDIA) NVIDIA Nemotron-3-Super-120B-A12B (NVIDIA) Gemma 4 (Google DeepMind) Step-3.5-Flash-Base (StepFun) Qwen3.5-0.8B (Alibaba / Qwen) Qwen3.5-2B (Alibaba / Qwen) Qwen3.5-4B (Alibaba / Qwen) Qwen3.5-35B-A3B-Base (Alibaba / Qwen) Rerank 4 Fast (Cohere) Rerank 4 Pro (Cohere) Amazon Nova 2 Lite (Amazon Web Services) Olmo 3 (Ai2) Grok 4.1 Thinking (SpaceXAI (formerly xAI)) Decision-1 (Microsoft) Gemini 3.8 Flash TTS (Google DeepMind) Gemini 3.8 Live Extended Thinking (Google DeepMind) Gemini 3.8 Live (Google DeepMind) Gemini 3.5 Transcribe (Google DeepMind) MAI Code 1.1 Flash (Microsoft) Gemini 3.5 Live Translate (Google DeepMind) Composer 2.5 (Anysphere) GPT-Realtime-2 (OpenAI) Gemini 3.1 Flash TTS (Google DeepMind) Gemini 3.1 Flash Live (Google DeepMind) Mistral Large 3 (Mistral AI) LFM2.5-VL-3B (Liquid AI) Nemotron-Labs-TwoTower-30B-A3B-Base-BF16 (NVIDIA) Gemma 4 12B (Google DeepMind) MAI-Thinking-1 (Microsoft) Magistral Small (Mistral AI) ZAYA1-8B (Zyphra) GPT-5.5 Instant (OpenAI) Granite 4.1 3B (IBM) Muse Spark (Meta AI) ERNIE 5.0 (Baidu AI)
Model B GPT-6.1 Terra (OpenAI) Kimi K3.1 Pro (Moonshot AI) Cohere Rerank 5 (Cohere) Mercury 3 (Inception) MiniMax M3.1 Flash (MiniMax) DeepSeek V4.1 Flash 1017 (DeepSeek) Glimpse (Elorian) Moondream 3.1 (Moondream) Muse Spark Open (Meta AI) Moondream 3 Preview (Moondream) Nemotron 3.5 Super VL (NVIDIA) ScaleDown: Summarize (ScaleDown) LFM2.5-24B-A2B (Liquid AI) Pareto 26.9 (Unbiased) Muse Wonder (Meta AI) Claude Haiku 5.5 (Anthropic) Mistral Large 4 (Mistral AI) Nano Banana 2.1 (Google DeepMind) Ling 3.1 Flash (inclusionAI) Gemini 4 Argon (Google DeepMind) GPT-6.1 Sol (OpenAI) Claude Sonnet 5.5 (Anthropic) Perceptron Mk1.5 (Perceptron) Ember-1 (Fireworks AI) Aion 3.5 (Aion Labs) Aion 3.5 Mini (Aion Labs) GLM-5.3-Prime (Zhipu AI) Solar Mini 4 (Upstage) Claude Opus 5.5 (Anthropic) GPT-6 Luna (OpenAI) GPT-6 Sol (OpenAI) MiMo-V2.6-Pro-UltraSpeed (Xiaomi) MiMo-V2.6-Pro (Xiaomi) Grok 4.7 (SpaceXAI (formerly xAI)) MiMo-V2.6-Flash (Xiaomi) Step 5 Preview (StepFun) Qwen3.8-Omni-Flash (Alibaba / Qwen) Ternary Bonsai 2 27B (PrismML) GLM-5.3-FlashX (Zhipu AI) Schematron V2 Small (Inference.net) Schematron V2 Turbo (Inference.net) Fugu Ultra v2 (Sakana AI) Sakana: Fugu Max (Sakana AI) DeepSeek-V4.1-Flash (DeepSeek) Nex-N2.5-mini (Nex-AGI) Nex-N2.5-Pro (Nex-AGI) Mercury 2.5 (Inception) Ling 3.0 Flash VL (inclusionAI) GPT-6 Astra (OpenAI) Meta: Muse Spark 1.3 Contributor (Meta AI) Gemini 3.8 Flash (Google DeepMind) Meta: Muse Spark 1.3 (Meta AI) Claude Fable 5.1 (Anthropic) Tencent Hy4 Preview (Tencent) GLM-5.3-Flash (Zhipu AI) Qwen3.8 Flash (Alibaba / Qwen) Granite 4.2 8B (IBM) DeepSeek V4 Flash Vision Exp (DeepSeek) Hy-MT2-30B-A3B (Tencent) Qwen3.8-27B (Alibaba / Qwen) GLM-5.3 (Zhipu AI) Gemini 3.7 Flash (Google DeepMind) DeepSeek V4 Pro 0813 (DeepSeek) Qwen3.8-2.4T-A95B (Alibaba / Qwen) Grok 4.6 (SpaceXAI (formerly xAI)) Solar Pro 4 (Upstage) Sakana Namazu (Sakana AI) Nemotron 3.5 Lightning (NVIDIA) Seed-2.0-Code (ByteDance) Seed 2.1 Turbo (ByteDance) Muse Glimmer (Meta AI) Meta Muse Glimmer 30B (Meta AI) Muse Spark 1.2 (Meta AI) LFM2.5-2.6B (Liquid AI) Qwen3.8-Max (Alibaba / Qwen) DeepSeek V4 Flash 0731 (DeepSeek) Inkling Small (Thinking Machines Lab) Qwen3.7 Flash (Alibaba / Qwen) Claude Opus 5 (Anthropic) Ling-3.0-flash (inclusionAI) Laguna S 2.1 (Poolside) LongCat 2.0 (Meituan) Kimi K3 (Moonshot AI) Inkling (Thinking Machines Lab) Muse Spark 1.1 (Meta AI) GPT-5.6 Luna Pro (OpenAI) GPT-5.6 Luna (OpenAI) GPT-5.6 Terra Pro (OpenAI) GPT-5.6 Terra (OpenAI) GPT-5.6 Sol (OpenAI) GPT-5.6 Sol Pro (OpenAI) Grok 4.5 (SpaceXAI (formerly xAI)) Aion-3.0-Mini (Aion Labs) Aion-3.0 (Aion Labs) Hy3 (Tencent) Laguna XS 2.1 (Poolside) Gemini 3.1 Flash Lite Image (Google DeepMind) Claude Sonnet 5 (Anthropic) Fugu Ultra (Sakana AI) Gemini 3.1 Flash Image (Google DeepMind) Nano Banana Pro (Gemini 3 Pro Image) (Google DeepMind) GLM-5.2 (Zhipu AI) Kimi K2.7 Code (Moonshot AI) Claude Fable 5 (Anthropic) Nemotron 3.5 Content Safety (NVIDIA) Nemotron 3 Ultra (NVIDIA) Qwen3.7 Plus (Alibaba / Qwen) MiniMax-M3 (MiniMax) Claude Opus 4.8 (Anthropic) Step-3.7-Flash (StepFun) Hy-MT2-7B (Tencent) Hy-MT2-1.8B (Tencent) Gemini 3.5 Flash (Google DeepMind) Qwen3.7 Max (Alibaba / Qwen) Command A+ (Cohere) Grok Build 0.1 (SpaceXAI (formerly xAI)) Gemini Omni Flash (Google DeepMind) Perceptron Mk1 (Perceptron) Ring-2.6-1T (inclusionAI) Gemini 3.1 Flash Lite (Google DeepMind) Laguna M.1 (Poolside) Mistral Medium 3.5 (Mistral AI) Qwen3.6 Flash (Alibaba / Qwen) GPT-5.5 Pro (OpenAI) GPT-5.5 (OpenAI) Qwen3.6 27B (Alibaba / Qwen) DeepSeek V4 Pro (DeepSeek) DeepSeek V4 Flash (DeepSeek) GPT-5.4 Image 2 (OpenAI) Qwen3.6 Max Preview (Alibaba / Qwen) Kimi K2.6 (Moonshot AI) Grok 4.3 (SpaceXAI (formerly xAI)) Claude Opus 4.7 (Anthropic) Qwen3.6 35B A3B (Alibaba / Qwen) Gemma 4 26B A4B (Google DeepMind) GLM-5V-Turbo (Zhipu AI) Qwen 3.6 Plus (Alibaba / Qwen) Gemma 4 31B (Google DeepMind) Trinity Large Thinking (Arcee AI) Grok 4.20 Multi-Agent (SpaceXAI (formerly xAI)) Reka Edge (Reka) MiMo-V2-Omni (Xiaomi) MiMo-V2-Pro (Xiaomi) GPT-5.4 Nano (OpenAI) GPT-5.4 mini (OpenAI) Mistral Small 4 (Mistral AI) MiniMax M2.7 (MiniMax) GLM 5 Turbo (Zhipu AI) Nemotron 3 Super (NVIDIA) GPT-5.4 (OpenAI) Qwen3.5-9B (Alibaba / Qwen) Qwen3.5-27B (Alibaba / Qwen) Qwen3.5-35B-A3B (Alibaba / Qwen) Qwen3.5-Flash (Alibaba / Qwen) Mercury 2 (Inception) Gemini 3.1 Pro (Google DeepMind) Grok 4.20 (SpaceXAI (formerly xAI)) GLM-5.1 (Zhipu AI) Claude Sonnet 4.6 (Anthropic) Qwen3.5 397B A17B (Alibaba / Qwen) Seed-2.0-Mini (ByteDance) Seed-2.0-Lite (ByteDance) Doubao Seed 2.0 (ByteDance) MiniMax-M2.5 (MiniMax) GLM 5 (Zhipu AI) GPT-5.3-Codex (OpenAI) Claude Opus 4.6 (Anthropic) Step-3.5-Flash (StepFun) Kimi K2.5 (Moonshot AI) GLM-4.7 (Zhipu AI) Gemini 3 Flash (Google DeepMind) GPT-5.2 (OpenAI) Claude Opus 4.5 (Anthropic) Kimi K2 Thinking (Moonshot AI) MiniMax-M2 (MiniMax) Nanbeige4.3-33B (Nanbeige) Raon-1 mini preview (Krafton) MOSS-Image (Openmoss) Ling-3.0-flash-Sante (inclusionAI) Sage-A1.5 (Shanghai AI Lab) OrcaCyber Nano 1.0 (OrcaRouter) Reflection Beam 1.0 (Reflection AI) NVIDIA Nemotron 3.5 Super VL 120B-A12B NVFP4 (NVIDIA) Beam (Reflection AI) Clef (Cloudflare) Space Bunny Alpha (Undisclosed (stealth models)) MiMo-V2.6-Pro-RL (Xiaomi) MiMo-V2.6-Flash-RL (Xiaomi) FrogNano-4B-2609 (Microsoft) AliceAI-Foundation-80B-A3B-Base (Yandex) Xing4.0-29B-A4B (China Telecom AI (TeleAI)) Union Alpha (Undisclosed (stealth models)) Iris-pro (Allspark) Atria Dawn (Shanghai AI Lab) Granite Time Series PatchTST-FM-r2 (IBM) Ling 3.0 Flash Fin (inclusionAI) Qwen3.8-Flash-Next (Alibaba / Qwen) IBM Granite 4.2 (30B) (IBM) Granite 4.2 (IBM) Ox Alpha (OpenRouter) MiniMax Music 3 (MiniMax) Shieldstral (Mistral AI) K-EXAONE 2.0 (750B-A37B) (LG AI Research) LFM2.5-Encoder (Liquid AI) Qwen-Image-3.0 (Alibaba / Qwen) Nemotron-3-Embed-1B-BF16 (NVIDIA) Nemotron 3 Embed 8B BF16 (NVIDIA) KAT-Coder-Air V2.5 (Kwaipilot) KAT-Coder-Pro V2.5 (Kwaipilot) Audex-30B-A3B (NVIDIA) NVIDIA Nemotron-Labs-3-Puzzle-75B-A9B-NVFP4 (NVIDIA) Leanstral 1.5 (Mistral AI) LFM2.5-230M (Liquid AI) DeepSeek-V4-Fable (DeepSeek) Qwen-AgentWorld-35B-A3B (Alibaba / Qwen) Unlimited-OCR (Baidu AI) North Mini Code (Cohere) FastContext-1.0-4B-SFT (Microsoft) DiffusionGemma (Google DeepMind) North Mini Code 1.0 (Cohere) Nex-N2-Pro (Nex-AGI) Gemma 4 12B Unified (Google DeepMind) Mellum2-12B-A2.5B-Thinking (JetBrains) Cosmos3-Nano (NVIDIA) Cosmos 3 Super (NVIDIA) Cosmos 3 Super Image2Video (NVIDIA) Cosmos3-Super-Text2Image (NVIDIA) Voxtral Small (Mistral AI) LocateAnything-3B (NVIDIA) Fara1.5-27B (Microsoft) Fara-7B (Microsoft) Granite Embedding 97M Multilingual R2 (IBM) Baidu Qianfan: CoBuddy (Baidu AI) Granite Speech 4.1 2B (IBM) Granite 4.1 30B (IBM) Granite 4.1 8B (IBM) Granite Embedding 311M Multilingual R2 (IBM) Laguna XS.2 (Poolside) Nemotron 3 Nano Omni 30B-A3B-Reasoning (NVIDIA) Owl Alpha (OpenRouter) Nemotron-3-Nano-Omni-30B-A3B (NVIDIA) Ling-2.6-1T (inclusionAI) Trinity Large Preview (Arcee AI) OpenAI Privacy Filter (OpenAI) Gemma 4 E2B VLA (Google DeepMind) Pareto Code Router (OpenRouter) Ling-2.6-flash (inclusionAI) LFM2.5-VL-450M (Liquid AI) HY-Embodied-0.5 MoT-2B (Tencent) Harrier (27B) (Microsoft) Amazon Nova 2 Sonic (Amazon Web Services) Gemma 4 E4B Instruction-Tuned (Google DeepMind) Gemma 4 31B Dense (Google DeepMind) Gemma 4 E2B (Google DeepMind) Harrier-OSS v1 270M (Microsoft) Qwen3.5-Omni (Alibaba / Qwen) Google Lyria 3 Clip Preview (Google DeepMind) KAT-Coder-Pro V2 (Kwaipilot) gpt-oss-puzzle-88B (NVIDIA) Qwen3.5-Max-Preview (Alibaba / Qwen) RakutenAI 3.0 (Rakuten) NVIDIA Nemotron-3-Nano-4B-GGUF (NVIDIA) NVIDIA Nemotron-3-Super-120B-A12B (NVIDIA) Gemma 4 (Google DeepMind) Step-3.5-Flash-Base (StepFun) Qwen3.5-0.8B (Alibaba / Qwen) Qwen3.5-2B (Alibaba / Qwen) Qwen3.5-4B (Alibaba / Qwen) Qwen3.5-35B-A3B-Base (Alibaba / Qwen) Rerank 4 Fast (Cohere) Rerank 4 Pro (Cohere) Amazon Nova 2 Lite (Amazon Web Services) Olmo 3 (Ai2) Grok 4.1 Thinking (SpaceXAI (formerly xAI)) Decision-1 (Microsoft) Gemini 3.8 Flash TTS (Google DeepMind) Gemini 3.8 Live Extended Thinking (Google DeepMind) Gemini 3.8 Live (Google DeepMind) Gemini 3.5 Transcribe (Google DeepMind) MAI Code 1.1 Flash (Microsoft) Gemini 3.5 Live Translate (Google DeepMind) Composer 2.5 (Anysphere) GPT-Realtime-2 (OpenAI) Gemini 3.1 Flash TTS (Google DeepMind) Gemini 3.1 Flash Live (Google DeepMind) Mistral Large 3 (Mistral AI) LFM2.5-VL-3B (Liquid AI) Nemotron-Labs-TwoTower-30B-A3B-Base-BF16 (NVIDIA) Gemma 4 12B (Google DeepMind) MAI-Thinking-1 (Microsoft) Magistral Small (Mistral AI) ZAYA1-8B (Zyphra) GPT-5.5 Instant (OpenAI) Granite 4.1 3B (IBM) Muse Spark (Meta AI) ERNIE 5.0 (Baidu AI)
Model C (optional) + Add a third model
Orange values indicate the highest score. Benchmark data from official model cards and technical reports. Full leaderboard →