GLM-5.2 Zhipu AIN/A N/A 1.0M 131K compare (6) 51.1 #1 68.8 #1 MiniMax M3 MiniMaxN/A N/A 1.0M 512K compare (6) 44.4 #2 58.6 #7 DeepSeek V4 Pro DeepSeekN/A N/A 1.0M 384K compare (9) 44.3 #3 59.4 #5 Kimi K2.6 Moonshot AI (Kimi)N/A N/A 262K 262K compare (9) 44.2 #4 61.8 #2 MiMo V2.5 Pro XiaomiN/A N/A 1.1M 131K compare (2) 42.2 #5 60.2 #4 Kimi K2.7 Code Moonshot AI (Kimi)N/A N/A 262K 33K compare (6) 41.9 #6 60.8 #3 Hy 3 TencentN/A N/A 262K 128K compare (2) 41.2 #7 58.8 #6 Inkling ThinkingmachinesN/A N/A 1.0M N/A compare (3) 40.7 #8 52.1 #12 DeepSeek V4 Flash DeepSeekN/A N/A 1.0M 384K compare (11) 40.3 #9 56.2 #8 GLM-5.1 Zhipu AIN/A N/A 205K 131K compare (7) 40.2 #10 55.8 #9 GLM-5 Zhipu AIN/A N/A 205K 131K compare (9) 39.5 #11 N/A MiniMax M2.7 MiniMaxN/A N/A 1.0M 131K compare (6) 38.1 #12 52.6 #11 Nemotron 3 Ultra 550B A55B NVIDIAN/A N/A 1.0M 66K compare (4) 37.8 #13 49.3 #14 Qwen3.6 27B AlibabaN/A N/A 262K 131K compare (5) 37.1 #14 53.7 #10 Kimi K2.5 Moonshot AI (Kimi)N/A N/A 262K 98K compare (12) 35.4 #15 46.8 #17 Qwen3.5 27B AlibabaN/A N/A 262K 66K compare (3) 33.8 #16 N/A Qwen3.5 397B A17B AlibabaN/A N/A 262K 66K compare (6) 33.7 #17 48.2 #15 MiniMax M2.5 MiniMaxN/A N/A 1.0M 197K compare (9) 33.7 #18 N/A GLM-4.7 Zhipu AIN/A N/A 205K 131K compare (13) 33.7 #19 45.3 #19 Kimi K2 Thinking Moonshot AI (Kimi)N/A N/A 262K 100K compare (13) 32.7 #20 N/A Qwen3.5 122B A10B AlibabaN/A N/A 262K 66K compare (4) 32.3 #21 45.7 #18 Qwen3.6 35B A3B AlibabaN/A N/A 262K 16K compare (6) 31.6 #22 41.9 #20 MiniMax M2.1 MiniMaxN/A N/A 1.0M 131K compare (8) 31.4 #23 N/A Step 3.7 Flash StepFunN/A N/A 262K 256K compare (2) 30.3 #24 39.6 #21 Qwen3.5 35B A3B AlibabaN/A N/A 262K 66K compare (3) 29.3 #25 N/A MiniMax M2 MiniMaxN/A N/A 205K 131K compare (9) 28.3 #26 N/A InclusionAI Ling 2.6 1T InclusionaiN/A N/A 262K 33K compare (2) 26.1 #27 N/A Step 3.5 Flash StepFunN/A N/A 262K 66K compare (2) 25.5 #28 N/A MiMo V2 Flash XiaomiN/A N/A 262K 32K compare (2) 24.7 #30 49.8 #13 DeepSeek V3.2 DeepSeekN/A N/A 164K 66K compare (13) 24.7 #29 N/A GPT OSS 120B OpenAIN/A N/A 131K 131K compare (24) 23.8 #31 30.4 #23 GLM-4.6 Zhipu AIN/A N/A 205K 131K compare (10) 23.0 #32 N/A GLM-4.7 Flash Zhipu AIN/A N/A 203K 131K compare (4) 22.9 #33 N/A DeepSeek V3.2 Speciale DeepSeekN/A N/A 164K N/A compare (1) 22.2 #34 N/A Qwen3.5 9B AlibabaN/A N/A 262K 33K compare (2) 21.4 #35 28.7 #24 DeepSeek V3.1 Terminus DeepSeekN/A N/A 164K 66K compare (6) 21.4 #36 N/A Qwen3 Coder Next AlibabaN/A N/A 262K 8K compare (5) 21.1 #37 36.2 #22 DeepSeek V3.1 DeepSeekN/A N/A 164K 128K compare (14) 21.0 #38 N/A GLM-4.5 Zhipu AIN/A N/A 131K 98K compare (7) 19.5 #39 N/A Kimi K2 Moonshot AI (Kimi)N/A N/A 262K 100K compare (4) 19.4 #40 47.1 #16 DeepSeek R1 DeepSeekN/A N/A 164K 66K compare (13) 18.5 #41 24.6 #26 Arcee Trinity Large Thinking Arcee AIN/A N/A 262K 80K compare (2) 18.2 #42 25.8 #25 Qwen3 Coder 480B A35B Instruct AlibabaN/A N/A 262K 66K compare (8) 18.0 #43 N/A MiniMax M1 80k MiniMaxN/A N/A 1.0M 40K compare (3) 17.7 #44 N/A GLM-4.5 Air Zhipu AIN/A N/A 131K 98K compare (7) 16.5 #45 N/A DeepSeek V3 324 DeepSeekN/A N/A 164K 16K compare (12) 15.4 #46 21.2 #28 GPT OSS 20B OpenAIN/A N/A 131K 131K compare (18) 14.9 #47 20.7 #29 Qwen3 VL 235B A22B Instruct AlibabaN/A N/A 262K 129K compare (7) 14.3 #48 N/A Llama 4 Maverick MetaN/A N/A 1.0M 16K compare (6) 14.3 #49 16.3 #30 DeepSeek V3 DeepSeekN/A N/A 164K 82K compare (11) 14.2 #50 23.0 #27 Qwen3 Next 80B A3B Instruct AlibabaN/A N/A 262K 66K compare (10) 13.7 #51 N/A Qwen3 Coder 30B A3B Instruct AlibabaN/A N/A 262K 33K compare (5) 13.6 #52 N/A QwQ 32B AlibabaN/A N/A 131K 16K compare (6) 13.4 #53 N/A Qwen3 VL 32B Instruct AlibabaN/A N/A 131K 33K compare (3) 11.1 #54 N/A GLM-4.6V Zhipu AIN/A N/A 131K 33K compare (4) 11.0 #56 N/A DeepSeek R1 Distill Qwen 32B DeepSeekN/A N/A 131K 32K compare (6) 11.0 #55 N/A Qwen3 235B A22B Instruct AlibabaN/A N/A 262K 16K compare (12) 10.9 #57 N/A Qwen3 VL 30B A3B Instruct AlibabaN/A N/A 262K 33K compare (5) 10.0 #58 N/A Llama 4 Scout MetaN/A N/A 1.3M 16K compare (4) 10.0 #59 8.2 #32 DeepSeek R1 Distill Llama 70B DeepSeekN/A N/A 131K 8K compare (10) 9.9 #60 N/A Qwen2.5 72B Instruct AlibabaN/A N/A 131K 16K compare (6) 9.6 #61 N/A Llama 3.3 70B Instruct MetaN/A N/A 131K 128K compare (21) 9.4 #62 11.9 #31 Qwen3 30B A3B AlibabaN/A N/A 131K 20K compare (7) 9.1 #63 N/A Qwen3 VL 8B Instruct AlibabaN/A N/A 262K 33K compare (4) 8.4 #64 N/A Llama 3.1 Nemotron 70B Instruct NVIDIAN/A N/A 131K 16K compare (2) 7.6 #66 N/A Llama 3.1 8B Instruct MetaN/A N/A 200K 128K compare (20) 7.6 #65 5.4 #33 Nemotron Nano 3 30B A3B NVIDIAN/A N/A 262K 228K compare (3) 7.4 #67 N/A Qwen2.5 Coder 32B Instruct AlibabaN/A N/A 131K 4K compare (8) 7.1 #68 N/A GLM-4.5V Zhipu AIN/A N/A 131K 32K compare (6) 7.0 #69 N/A Qwen3 30B A3B Instruct AlibabaN/A N/A 262K 32K compare (4) 6.8 #70 N/A Llama 3.1 70B Instruct MetaN/A N/A 131K 16K compare (12) 6.8 #71 N/A Hermes 3 Llama 3.1 70B Nous ResearchN/A N/A 131K 16K compare (3) 5.1 #72 N/A Phi-4 MicrosoftN/A N/A 16K 16K compare (3) 4.9 #73 N/A Mixtral 8x22B Instruct Mistral AIN/A N/A 66K 2K compare (6) 4.4 #74 N/A Llama 3.2 3B Instruct MetaN/A N/A 131K 32K compare (10) 4.2 #75 N/A Llama 3 70B Instruct MetaN/A N/A 131K 8K compare (8) 3.5 #76 N/A Mistral 7B Instruct Mistral AIN/A N/A 127K 16K compare (8) 2.1 #77 N/A Llama 3 8B Instruct MetaN/A N/A 32K 8K compare (8) 1.2 #78 N/A Llama 3.2 1B Instruct MetaN/A N/A 128K 16K compare (6) 1.1 #79 N/A Arctic 2 Embed L SnowflakeN/A N/A 8K N/A compare (1) N/A N/A AutoGLM Phone 9B Multilingual Zhipu AIN/A N/A 66K N/A compare (2) N/A N/A BGE Base EN V1.5 BaaiN/A N/A 154K N/A compare (2) N/A N/A BGE Large EN V1.5 BaaiN/A N/A 512 N/A compare (1) N/A N/A BGE-M3 BaaiN/A N/A 60K N/A compare (3) N/A N/A Code Llama 7B Instruct MetaN/A N/A 16K 16K compare (1) N/A N/A DeepSeek R1 528 DeepSeekN/A N/A 164K 33K compare (12) N/A N/A DeepSeek V3 0324 DeepSeekN/A N/A 164K 66K compare (1) N/A N/A DeepSeek V3.1 DeepSeekN/A N/A 164K 33K compare (1) N/A N/A DeepSeek V3.2 Exp DeepSeekN/A N/A 164K 66K compare (4) N/A N/A DeepSeek-OCR DeepSeekN/A N/A 8K 8K compare (4) N/A N/A E5 Base V2 MicrosoftN/A N/A N/A N/A compare (1) N/A N/A ERNIE 4.5 21B A3B BaiduN/A N/A 120K 8K compare (1) N/A N/A ERNIE 4.5 VL 28B A3B BaiduN/A N/A 30K 8K compare (1) N/A N/A FLUX 2 Dev Black Forest LabsN/A N/A N/A N/A compare (1) N/A N/A FLUX Klein 2 4B Black Forest LabsN/A N/A N/A N/A compare (1) N/A N/A FLUX Klein 2 9B Black Forest LabsN/A N/A N/A N/A compare (1) N/A N/A FLUX.1 Dev Black Forest LabsN/A N/A 4K 4K compare (1) N/A N/A FLUX.1 Schnell Black Forest LabsN/A N/A 4K 4K compare (2) N/A N/A Gemma 2 27B Instruct GoogleN/A N/A 8K 8K compare (2) N/A N/A Gemma 3 12B Instruct GoogleN/A N/A 131K 16K compare (6) N/A N/A Gemma 3 27B Instruct GoogleN/A N/A 262K 131K compare (8) N/A N/A Gemma 3 4B Instruct GoogleN/A N/A 131K 16K compare (3) N/A N/A Gemma 3N E4B IT GoogleN/A N/A 33K 2K compare (2) N/A N/A Gemma 4 26B A4B IT GoogleN/A N/A 262K 131K compare (6) N/A N/A Gemma 4 31B IT GoogleN/A N/A 262K 131K compare (6) N/A N/A GPT OSS 20B Safeguard OpenAIN/A N/A 131K 66K compare (5) N/A N/A GTE Base AlibabaN/A N/A 512 N/A compare (1) N/A N/A GTE Large AlibabaN/A N/A 512 N/A compare (1) N/A N/A Hermes 2 Pro Llama 3 8B Nous ResearchN/A N/A 8K 8K compare (1) N/A N/A Hermes 4 70B Nous ResearchN/A N/A 131K N/A compare (2) N/A N/A Hermes Llama2 13B Nous ResearchN/A N/A 4K 4K compare (1) N/A N/A Kimi K2 Instruct Moonshot AI (Kimi)N/A N/A 262K 33K compare (9) N/A N/A L3 70B Euryale V2.1 Sao10KN/A N/A 8K 8K compare (1) N/A N/A L3 Lunaris 8B Sao10KN/A N/A 8K 8K compare (3) N/A N/A L3 Stheno 3.2 8B Sao10KN/A N/A 8K N/A compare (2) N/A N/A L3.3 70B Euryale V2.3 Sao10KN/A N/A 131K N/A compare (1) N/A N/A L3.3 Euryale 70B Sao10KN/A N/A 131K 16K compare (1) N/A N/A Llama 3.3 Nemotron 1.5 Super 49B NVIDIAN/A N/A 131K 16K compare (1) N/A N/A Llama 4 17B Maverick Instruct MetaN/A N/A 1.0M 16K compare (9) N/A N/A Llama 4 17B Scout Instruct MetaN/A N/A 10.0M 16K compare (12) N/A N/A LlamaGuard 3 8B MetaN/A N/A 131K 16K compare (4) N/A N/A LlamaGuard 4 12B MetaN/A N/A 1.0M 16K compare (4) N/A N/A Magnum V4 72B AnthraciteN/A N/A 16K 2K compare (1) N/A N/A MiMo V2.5 XiaomiN/A N/A 1.1M 131K compare (2) N/A N/A Mistral 7B Mistral AIN/A N/A 33K 8K compare (5) N/A N/A Mistral Small 3.1 24B Instruct Mistral AIN/A N/A 131K N/A compare (3) N/A N/A Multilingual E5 Large MicrosoftN/A N/A N/A N/A compare (1) N/A N/A Multilingual E5 Large Instruct MicrosoftN/A N/A N/A N/A compare (2) N/A N/A MythoMax L2 13B GrypheN/A N/A 8K 4K compare (4) N/A N/A Phi-4 Mini Instruct MicrosoftN/A N/A 131K 4K compare (2) N/A N/A Qwen2.5 7B Instruct AlibabaN/A N/A 33K 33K compare (4) N/A N/A Qwen2.5 VL 72B Instruct AlibabaN/A N/A 131K 128K compare (6) N/A N/A Qwen3 14B AlibabaN/A N/A 131K 16K compare (6) N/A N/A Qwen3 235B A22B AlibabaN/A N/A 262K 131K compare (8) N/A N/A Qwen3 235B A22B Thinking AlibabaN/A N/A 262K 33K compare (9) N/A N/A Qwen3 32B AlibabaN/A N/A 131K 41K compare (15) N/A N/A Qwen3 8B AlibabaN/A N/A 131K 20K compare (6) N/A N/A Qwen3 Coder AlibabaN/A N/A 262K 67K compare (2) N/A N/A Qwen3 Embedding 8B AlibabaN/A N/A 41K 4K compare (5) N/A N/A Qwen3 Next 80B A3B Thinking AlibabaN/A N/A 262K 66K compare (10) N/A N/A Qwen3 VL 235B A22B Thinking AlibabaN/A N/A 262K 33K compare (5) N/A N/A Qwen3 VL 30B A3B Thinking AlibabaN/A N/A 262K 33K compare (4) N/A N/A Qwen3 VL 8B Thinking AlibabaN/A N/A 131K 33K compare (2) N/A N/A ReMM SLERP L2 13B Undi95N/A N/A 6K 4K compare (1) N/A N/A UI-TARS 1.5 7B ByteDanceN/A N/A 131K 2K compare (1) N/A N/A Whisper 3 Large OpenAIN/A N/A N/A N/A compare (1) N/A N/A Whisper 3 Large Turbo OpenAIN/A N/A N/A N/A compare (3) N/A N/A