| 1 | paraphrase-mpnet-base-v2sentence-transformers | Sentence Similarity | 4 | ▲1.1%(up) | 55 | 109M |
| 2 | Qwen2-1.5BQwen | Text Generation | 4 | ▲1.1%(up) | 108 | 1.5B |
| 3 | reformer-crime-and-punishmentgoogle | Text Generation | 4 | ▲1.4%(up) | 15 | — |
| 4 | Fara1.5-4Bmicrosoft | Image Text to Text | 4 | ▲5.2%(up) | 54 | 4.5B |
| 5 | Wan2.2-S2V-14BWan-AI | Image To Video | 4 | ▲1.5%(up) | 489 | 16.3B |
| 6 | gemma-4-26B-A4Bgoogle | Image Text to Text | 4 | ▲0.7%(up) | 437 | 26.5B |
| 7 | Gemma-4-26B-A4B-NVFP4nvidia | Text Generation | 4 | ▲3.2%(up) | 159 | 14.4B |
| 8 | Pixal3DComfy-Org | — | 4 | ▲34.9%(up) | 101 | — |
| 9 | edda-v0.2danish-foundation-models | Speech Recognition | 4 | — | 6 | 809M |
| 10 | Qwen3-TTS-12Hz-1.7B-VoiceDesignQwen | Text to Speech | 4 | ▲2.1%(up) | 436 | 1.9B |
| 11 | OptiMind-SFTmicrosoft | Text Generation | 4 | ▲8.6%(up) | 108 | 20.9B |
| 12 | resnet50.a1_in1ktimm | Image Classification | 4 | ▲0.1%(up) | 48 | 25.6M |
| 13 | Qwen2-Audio-7BQwen | Audio Text To Text | 4 | ▲0.9%(up) | 180 | 8.4B |
| 14 | MiniMax-H3-Acc-LoRAsalibaba-pai | Text to Video | 4 | ▲15.8%(up) | 266 | — |
| 15 | baguettotron-600mPleIAs | Text Generation | 4 | — | 31 | 608M |
| 16 | MiniMax-H3-Prompt-Rewriter-LoRA-Omnilightx2v | Text Generation | 4 | — | 54 | — |
| 17 | KolorsKwai-Kolors | Text to Image | 4 | ▲0.2%(up) | 866 | 2.6B |
| 18 | Qwen3.8-27B-GSQ3-NInferroofkid | Image Text to Text | 4 | — | 5 | — |
| 19 | switch-c-2048google | — | 4 | ▲1.1%(up) | 306 | — |
| 20 | EVIE-8Btencent | Visual Document Retrieval | 4 | — | 34 | 8.4B |
| 21 | Ternary-Bonsai-8B-ggufprism-ml | Text Generation | 4 | ▲13.6%(up) | 169 | — |
| 22 | FastVideo-FastH3-8-Step-V2-NVFP4-ComfyFastVideo | — | 4 | — | 4 | — |
| 23 | gemma-4-E4B-it-MLX-6bitlmstudio-community | Any to Any | 4 | ▲5.4%(up) | 8 | 8B |
| 24 | Qwen3.5-9B-abliterated-GGUFAbiray | — | 4 | ▲5.8%(up) | 16 | — |
| 25 | LFM2.5-350MLiquidAI | Text Generation | 4 | ▲2.7%(up) | 434 | 354M |
| 26 | gemma-4-E2B-it-qat-GGUFunsloth | Any to Any | 4 | ▲12.9%(up) | 90 | — |
| 27 | table-transformer-detectionmicrosoft | Object Detection | 4 | ▲0.2%(up) | 435 | 28.8M |
| 28 | Qwen-Image-2.1-FP8unsloth | Text to Image | 4 | — | 42 | — |
| 29 | NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4nvidia | Text Generation | 4 | ▲9.3%(up) | 358 | 303B |
| 30 | Qwen3.8-3.6-27B-blend-MLX-4bitJetBrains | Image Text to Text | 4 | — | 10 | 27.4B |
| 31 | GLM-5.3-Flash-ABLITERATED-FP8dealignai | — | 4 | ▲17.1%(up) | 22 | 321B |
| 32 | Spark-X2.5-4B-uncensored-GGUFInfinimindCreations | Text Generation | 4 | ▲91.1%(up) | 12 | — |
| 33 | Mathstral-7B-v0.1mistralai | — | 4 | ▲0.0%(up) | 251 | 7.2B |
| 34 | Qwen3-VL-32B-InstructQwen | Image Text to Text | 4 | ▲0.6%(up) | 246 | 33.4B |
| 35 | stable-cascadestabilityai | Text to Image | 4 | ▲0.2%(up) | 1,337 | — |
| 36 | G4-MeroMero-26B-A4Bzerofata | — | 4 | ▲2.2%(up) | 128 | 25.8B |
| 37 | DeepSeek-V4.1-Flash-GSQ-RCO-GGUFpfeifferj | Image Text to Text | 4 | — | 19 | — |
| 38 | resnet-50microsoft | Image Classification | 4 | ▲0.0%(up) | 524 | 25.6M |
| 39 | Wan2.1-T2V-14BWan-AI | Text to Video | 4 | ▲1.4%(up) | 1,574 | 14.3B |
| 40 | OpenELMapple | — | 4 | — | 1,455 | — |
| 41 | snowflake-arctic-embed-lSnowflake | Sentence Similarity | 4 | ▲10.7%(up) | 108 | 334M |
| 42 | gemma-4-12B-it-qat-q4_0-unquantizedgoogle | Any to Any | 4 | ▲0.2%(up) | 88 | 12B |
| 43 | Hemmingway-1-Heretic-MTP-V3-Final-GGUFbrainnxdomain | — | 4 | — | 17 | — |
| 44 | wav2vec2-large-xlsr-53-arabicjonatasgrosman | Speech Recognition | 4 | ▲1.3%(up) | 59 | — |
| 45 | gemma-4-31B-it-qat-q4_0-ggufgoogle | Image Text to Text | 4 | ▲11.2%(up) | 138 | — |
| 46 | acestep-v15-xl-turboACE-Step | Text to Audio | 4 | ▲4.5%(up) | 223 | 5B |
| 47 | Qwen2.5-14B-Instruct-1MQwen | Text Generation | 4 | ▲0.4%(up) | 344 | 14.8B |
| 48 | Qwen3-VL-2B-ThinkingQwen | Image Text to Text | 4 | ▲5.5%(up) | 121 | 2.1B |
| 49 | siglip2-base-patch16-224google | Zero-Shot Image Classification | 4 | ▲4.8%(up) | 149 | 375M |
| 50 | Phi-4-multimodal-instructmicrosoft | Speech Recognition | 4 | ▲0.6%(up) | 1,619 | 5.6B |