💡 There's a 342,822x difference between the highest and lowest energy use in this leaderboard.
Select Model Class
Sort
💡 There's a 62x difference between the highest and lowest energy use in this class.
| Model | Provider | GPU Energy (Wh) per 1k Queries | Score | Test Date |
|---|---|---|---|---|
| distilgpt2 | distilbert | 1.31 | ★★★★★ | Feb 25 |
| opt-125m | 1.94 | ★★★★★ | Feb 25 | |
| gpt2 | openai-community | 2.15 | ★★★★★ | Feb 25 |
| openai-gpt | openai-community | 2.70 | ★★★★★ | Feb 25 |
| gpt2-medium | openai-community | 3.96 | ★★★★★ | Feb 25 |
| Hunyuan-1.8B-Instruct | tencent | 4.20 | ★★★★★ | Dec 25 |
| OLMo-1B-hf | allenai | 5.59 | ★★★★★ | Feb 25 |
| gpt2-large | openai-community | 6.12 | ★★★★★ | Feb 25 |
| phi-1_5 | microsoft | 6.29 | ★★★★★ | Feb 25 |
| phi-1 | microsoft | 6.39 | ★★★★★ | Feb 25 |
| SmolLM-135M | HuggingFaceTB | 7.28 | ★★★★ | Feb 25 |
| SmolLM-1.7B | HuggingFaceTB | 7.64 | ★★★★ | Feb 25 |
| SmolLM2-1.7B-Instruct | HuggingFaceTB | 7.71 | ★★★★ | Feb 25 |
| opt-2.7b | 8.06 | ★★★★ | Feb 25 | |
| SmolLM-360M | HuggingFaceTB | 8.10 | ★★★★ | Feb 25 |
| gemma-2b | 8.68 | ★★★★ | Feb 25 | |
| gpt2-xl | openai-community | 8.92 | ★★★★ | Feb 25 |
| gemma-3-270m | 9.29 | ★★★★★ | Dec 25 | |
| phi-2 | microsoft | 9.86 | ★★★★ | Feb 25 |
| Qwen3-0.6B | Qwen | 11.83 | ★★★★★ | Dec 25 |
| gemma-3-1b-pt | 12.33 | ★★★★★ | Dec 25 | |
| falcon-7b | tiiuae | 15.88 | ★★★ | Feb 25 |
| Llama-2-7b-hf | meta-llama | 16.44 | ★★★ | Feb 25 |
| granite-4.0-micro | ibm-granite | 16.63 | ★★★★★ | Dec 25 |
| OLMo-1.7-7B-hf | allenai | 16.83 | ★★★ | Feb 25 |
| Meta-Llama-3-8B | meta-llama | 17.33 | ★★★ | Feb 25 |
| Meta-Llama-3.1-8B | meta-llama | 17.38 | ★★★ | Feb 25 |
| aya-23-8B | CohereForAI | 17.82 | ★★ | Feb 25 |
| SmolLM3-3B | HuggingFaceTB | 18.35 | ★★★★★ | Dec 25 |
| Phi-4-reasoning-plus | microsoft | 18.42 | ★★★★★ | Dec 25 |
| gemma-2-2b-it | 18.64 | ★★ | Feb 25 | |
| gemma-7b | 18.90 | ★★ | Feb 25 | |
| Starling-LM-7B-alpha | berkeley-nest | 19.01 | ★★ | Feb 25 |
| Mistral-7B-v0.1 | mistralai | 19.13 | ★★ | Feb 25 |
| xLAM-7b-r | Salesforce | 19.17 | ★★ | Feb 25 |
| Ministral-8B-Instruct-2410 | mistralai | 20.88 | ★★ | Feb 25 |
| Apertus-8B-Instruct-2509 | swiss-ai | 21.74 | ★★★★ | Dec 25 |
| Yi-1.5-9B | 01-ai | 23.00 | ★★ | Feb 25 |
| Mistral-Nemo-Instruct-2407 | mistralai | 23.19 | ★★★★ | Dec 25 |
| Llama-2-13b-hf | meta-llama | 26.10 | ★ | Feb 25 |
| Qwen2.5-Coder-14B | Qwen | 27.12 | ★★★★ | Dec 25 |
| phi-4 | microsoft | 28.74 | ★ | Feb 25 |
| gemma-3-4b-pt | 34.39 | ★★★★ | Dec 25 | |
| granite-4.0-h-tiny | ibm-granite | 81.82 | ★ | Dec 25 |
Select Model Class
Sort
💡 There's a 23x difference between the highest and lowest energy use in this class.
| Model | Provider | GPU Energy (Wh) per 1k Queries | Score | Test Date |
|---|---|---|---|---|
| Qwen3-0.6B | Qwen | 546.99 | ★★★★★ | Dec 25 |
| Hunyuan-1.8B-Instruct | tencent | 1,478.96 | ★★★★★ | Dec 25 |
| Phi-4-reasoning-plus | microsoft | 9,461.61 | ★★ | Dec 25 |
| gemma-3-4b-pt | 11,811.18 | ★ | Dec 25 | |
| SmolLM3-3B | HuggingFaceTB | 12,791.22 | ★ | Dec 25 |
Sort
💡 There's a 9x difference between the highest and lowest energy use in this task.
| Model | Provider | GPU Energy (Wh) per 1k Queries | Score | Test Date |
|---|---|---|---|---|
| mitsua-diffusion-one | Mitsua | 186.81 | ★★★★★ | Feb 25 |
| Counterfeit-V2.5 | gsdf | 187.37 | ★★★★★ | Feb 25 |
| sd-turbo | stabilityai | 189.86 | ★★★★★ | Feb 25 |
| openjourney | prompthero | 196.88 | ★★★★★ | Feb 25 |
| openjourney-v4 | prompthero | 203.37 | ★★★★★ | Feb 25 |
| LCM_Dreamshaper_v7 | SimianLuo | 322.04 | ★★★★★ | Feb 25 |
| sdxl-turbo | stabilityai | 385.57 | ★★★★★ | Feb 25 |
| stable-diffusion-2-1 | stabilityai | 534.47 | ★★★★ | Feb 25 |
| dreamlike-photoreal-2.0 | dreamlike-art | 581.33 | ★★★★ | Feb 25 |
| epiCPhotoGasm | Yntec | 586.57 | ★★★★ | Feb 25 |
| stable-cascade | stabilityai | 1,214.36 | ★★ | Feb 25 |
| stable-diffusion-xl-base-1.0 | stabilityai | 1,639.85 | ★ | Feb 25 |
Sort
💡 There's a 129x difference between the highest and lowest energy use in this task.
| Model | Provider | GPU Energy (Wh) per 1k Queries | Score | Test Date |
|---|---|---|---|---|
| distilbert-imdb | lvwerra | 0.22 | ★★★★★ | Feb 25 |
| distilbert-base-uncased-finetuned-sst-2-english | distilbert | 0.22 | ★★★★★ | Feb 25 |
| distilbert-base-multilingual-cased-sentiments-student | lxyuan | 0.22 | ★★★★★ | Feb 25 |
| electra-base-emotion | bhadresh-savani | 0.25 | ★★★★★ | Feb 25 |
| gpt2-imdb-sentiment-classifier | mnoukhov | 0.32 | ★★★★★ | Feb 25 |
| bert-base-multilingual-uncased-sentiment | nlptown | 0.36 | ★★★★★ | Feb 25 |
| gpt2-medium-finetuned-sst2-sentiment | michelecafagna26 | 0.80 | ★★★★★ | Feb 25 |
| bert-large-sst2 | Cheng98 | 0.92 | ★★★★★ | Feb 25 |
| t5-small | google-t5 | 2.22 | ★★★★★ | Feb 25 |
| t5-base | google-t5 | 4.13 | ★★★★★ | Feb 25 |
| t5-large | google-t5 | 7.92 | ★★★★ | Feb 25 |
| t5-3b | google-t5 | 11.68 | ★★★ | Feb 25 |
| t5-11b | google-t5 | 27.79 | ★ | Feb 25 |
Sort
💡 There's a 9x difference between the highest and lowest energy use in this task.
| Model | Provider | GPU Energy (Wh) per 1k Queries | Score | Test Date |
|---|---|---|---|---|
| resnet-18 | microsoft | 0.30 | ★★★★★ | Feb 25 |
| vit-tiny-patch16-224 | WinKawaks | 0.38 | ★★★★★ | Feb 25 |
| resnet-50 | microsoft | 0.39 | ★★★★★ | Feb 25 |
| mobilenet_v2_1.0_224 | 0.41 | ★★★★★ | Feb 25 | |
| convnext-tiny-224 | 0.45 | ★★★★★ | Feb 25 | |
| mobilevit-small | apple | 0.51 | ★★★★★ | Feb 25 |
| vit-base-patch16-224 | 0.58 | ★★★★★ | Feb 25 | |
| beit-base-patch16-224 | microsoft | 0.74 | ★★★★★ | Feb 25 |
| vit-base-patch32-384 | 0.76 | ★★★★★ | Feb 25 | |
| swinv2-tiny-patch4-window8-256 | microsoft | 0.77 | ★★★★ | Feb 25 |
| beit-base-patch16-224-pt22k-ft22k | microsoft | 0.78 | ★★★★ | Feb 25 |
| bit-50 | 0.83 | ★★★★ | Feb 25 | |
| swinv2-tiny-patch4-window16-256 | microsoft | 0.85 | ★★★★ | Feb 25 |
| convnext-large-224 | 0.97 | ★★★★ | Feb 25 | |
| vit-base-patch16-384 | 0.99 | ★★★★ | Feb 25 | |
| convnext-base-384-22k-1k | 1.12 | ★★★★ | Feb 25 | |
| vit-large-patch32-384 | 1.16 | ★★★★ | Feb 25 | |
| swinv2-large-patch4-window12-192-22k | microsoft | 1.29 | ★★★ | Feb 25 |
| swinv2-large-patch4-window12-192-22k | microsoft | 1.39 | ★★★ | Feb 25 |
| swin-base-patch4-window12-384 | microsoft | 1.53 | ★★★ | Feb 25 |
| efficientnet-b7 | 1.85 | ★★ | Feb 25 | |
| mit-b5 | nvidia | 2.49 | ★ | Feb 25 |
| vision-perceiver-conv | deepmind | 2.64 | ★ | Feb 25 |
Sort
💡 There's a 4x difference between the highest and lowest energy use in this task.
| Model | Provider | GPU Energy (Wh) per 1k Queries | Score | Test Date |
|---|---|---|---|---|
| pix2struct-base | 45.75 | ★★★★★ | Feb 25 | |
| blip-image-captioning-base | Salesforce | 47.12 | ★★★★★ | Feb 25 |
| blip-image-captioning-large | Salesforce | 48.51 | ★★★★★ | Feb 25 |
| git-base | microsoft | 74.71 | ★★★★ | Feb 25 |
| blip2-opt-2.7b | Salesforce | 101.34 | ★★★ | Feb 25 |
| blip2-flan-t5-xl | Salesforce | 102.15 | ★★★ | Feb 25 |
| vlrm-blip2-opt-2.7b | sashakunitsyn | 103.35 | ★★★ | Feb 25 |
| git-large-coco | microsoft | 170.32 | ★ | Feb 25 |
| blip2-opt-6.7b | Salesforce | 174.96 | ★ | Feb 25 |
| blip2-flan-t5-xxl | Salesforce | 179.93 | ★ | Feb 25 |
Sort
💡 There's a 44x difference between the highest and lowest energy use in this task.
| Model | Provider | GPU Energy (Wh) per 1k Queries | Score | Test Date |
|---|---|---|---|---|
| bart_summarizer_model | KipperDev | 0.78 | ★★★★★ | Feb 25 |
| distilbart-xsum-12-1 | sshleifer | 1.05 | ★★★★★ | Feb 25 |
| distilbart-cnn-12-6 | sshleifer | 1.48 | ★★★★★ | Feb 25 |
| bart-large-cnn | 2.04 | ★★★★★ | Feb 25 | |
| t5-small | google-t5 | 2.27 | ★★★★★ | Feb 25 |
| bigbird-pegasus-large-arxiv | 3.52 | ★★★★★ | Feb 25 | |
| t5-base | google-t5 | 4.23 | ★★★★★ | Feb 25 |
| led-large-book-summary | pszemraj | 4.29 | ★★★★★ | Feb 25 |
| t5-large | google-t5 | 8.68 | ★★★★ | Feb 25 |
| t5-3b | google-t5 | 13.24 | ★★★★ | Feb 25 |
| t5-11b | google-t5 | 34.81 | ★ | Feb 25 |
Sort
💡 There's a 140x difference between the highest and lowest energy use in this task.
| Model | Provider | GPU Energy (Wh) per 1k Queries | Score | Test Date |
|---|---|---|---|---|
| wav2vec2-base-960h | 0.62 | ★★★★★ | Feb 25 | |
| sew-d-base-plus-400k-ft-ls100h | asapp | 1.54 | ★★★★★ | Feb 25 |
| hubert-large-ls960-ft | 1.62 | ★★★★★ | Feb 25 | |
| mms-1b-all | 4.28 | ★★★★★ | Feb 25 | |
| whisper-tiny | openai | 13.44 | ★★★★★ | Feb 25 |
| distil-small.en | distil-whisper | 15.57 | ★★★★★ | Feb 25 |
| whisper-base | openai | 16.55 | ★★★★★ | Feb 25 |
| distil-medium.en | distil-whisper | 16.97 | ★★★★★ | Feb 25 |
| s2t-small-librispeech-asr | 17.20 | ★★★★★ | Feb 25 | |
| distil-large-v2 | distil-whisper | 21.52 | ★★★★ | Feb 25 |
| whisper-small | openai | 29.97 | ★★★★ | Feb 25 |
| whisper-medium | openai | 61.42 | ★★ | Feb 25 |
| whisper-large-v3 | openai | 86.68 | ★ | Feb 25 |
Sort
💡 There's a 8x difference between the highest and lowest energy use in this task.
| Model | Provider | GPU Energy (Wh) per 1k Queries | Score | Test Date |
|---|---|---|---|---|
| yolos-tiny | hustvl | 1.00 | ★★★★★ | Feb 25 |
| conditional-detr-resnet-50 | microsoft | 2.14 | ★★★★★ | Feb 25 |
| detr-resnet-101 | 2.53 | ★★★★ | Feb 25 | |
| detr-resnet-50-dc5 | 3.28 | ★★★★ | Feb 25 | |
| yolos-small | hustvl | 3.28 | ★★★★ | Feb 25 |
| detr-resnet-50-dc5 | 3.29 | ★★★★ | Feb 25 | |
| yolos-base | hustvl | 7.98 | ★ | Feb 25 |
Sort
💡 There's a 49x difference between the highest and lowest energy use in this task.
| Model | Provider | GPU Energy (Wh) per 1k Queries | Score | Test Date |
|---|---|---|---|---|
| GIST-all-MiniLM-L6-v2 | avsolatorio | 0.11 | ★★★★★ | Feb 25 |
| bge-small-en-v1.5 | BAAI | 0.19 | ★★★★★ | Feb 25 |
| gte-base | thenlper | 0.25 | ★★★★★ | Feb 25 |
| bge-base-en-v1.5 | BAAI | 0.25 | ★★★★★ | Feb 25 |
| e5-base-v2 | intfloat | 0.25 | ★★★★★ | Feb 25 |
| GIST-Embedding-v0 | avsolatorio | 0.26 | ★★★★★ | Feb 25 |
| nomic-embed-text-v1 | nomic-ai | 0.38 | ★★★★★ | Feb 25 |
| ember-v1 | llmrails | 0.58 | ★★★★★ | Feb 25 |
| GIST-large-Embedding-v0 | avsolatorio | 0.58 | ★★★★★ | Feb 25 |
| mxbai-embed-large-v1 | mixedbread-ai | 0.59 | ★★★★★ | Feb 25 |
| gte-large | thenlper | 0.59 | ★★★★★ | Feb 25 |
| UAE-Large-V1 | WhereIsAI | 0.60 | ★★★★★ | Feb 25 |
| e5-large-v2 | intfloat | 0.60 | ★★★★★ | Feb 25 |
| multilingual-e5-large-instruct | intfloat | 0.72 | ★★★★★ | Feb 25 |
| udever-bloom-1b1 | izhx | 0.97 | ★★★★★ | Feb 25 |
| udever-bloom-3b | izhx | 2.15 | ★★★ | Feb 25 |
| udever-bloom-7b1 | izhx | 4.70 | ★ | Feb 25 |
| GritLM-7B | GritLM | 5.03 | ★ | Feb 25 |
| SFR-Embedding-Mistral | Salesforce | 5.22 | ★ | Feb 25 |
Sort
💡 There's a 3,215x difference between the highest and lowest energy use in this task.
| Model | Provider | GPU Energy (Wh) per 1k Queries | Score | Test Date |
|---|---|---|---|---|
| bert-tiny-finetuned-squadv2 | mrm8488 | 0.06 | ★★★★★ | Feb 25 |
| dynamic_tinybert | Intel | 0.21 | ★★★★★ | Feb 25 |
| question_answering_v2 | Falconsai | 0.23 | ★★★★★ | Feb 25 |
| distilbert-base-cased-distilled-squad | distilbert | 0.24 | ★★★★★ | Feb 25 |
| minilm-uncased-squad2 | deepset | 0.25 | ★★★★★ | Feb 25 |
| electra-base-squad2 | deepset | 0.47 | ★★★★★ | Feb 25 |
| deberta-v3-base-squad2 | deepset | 0.81 | ★★★★★ | Feb 25 |
| mdeberta-v3-base-squad2 | timpal0l | 0.86 | ★★★★★ | Feb 25 |
| mobilebert-uncased-squad-v2 | csarron | 0.94 | ★★★★★ | Feb 25 |
| bert-large-uncased-whole-word-masking-squad2 | deepset | 1.05 | ★★★★★ | Feb 25 |
| longformer-base-4096-finetuned-squadv2 | mrm8488 | 1.79 | ★★★★★ | Feb 25 |
| deberta-v3-large-squad2 | deepset | 1.88 | ★★★★★ | Feb 25 |
| longformer-large-4096-finetuned-triviaqa | allenai | 3.97 | ★★★★★ | Feb 25 |
| t5-small | google-t5 | 20.64 | ★★★★★ | Feb 25 |
| t5-base | google-t5 | 34.36 | ★★★★★ | Feb 25 |
| t5-large | google-t5 | 72.28 | ★★★ | Feb 25 |
| t5-3b | google-t5 | 87.70 | ★★★ | Feb 25 |
| t5-11b | google-t5 | 178.13 | ★ | Feb 25 |
Sort
💡 There's a 342,822x difference between the highest and lowest energy use in this leaderboard.
| Model | Provider | GPU Energy (Wh) per 1k Queries | Score | Task | Test Date |
|---|---|---|---|---|---|
| bert-tiny-finetuned-squadv2 | mrm8488 | 0.06 | ★★★★★ | Extractive QA ❔ | Feb 25 |
| GIST-all-MiniLM-L6-v2 | avsolatorio | 0.11 | ★★★★★ | Sentence Similarity 📚 | Feb 25 |
| bge-small-en-v1.5 | BAAI | 0.19 | ★★★★★ | Sentence Similarity 📚 | Feb 25 |
| dynamic_tinybert | Intel | 0.21 | ★★★★★ | Extractive QA ❔ | Feb 25 |
| distilbert-imdb | lvwerra | 0.22 | ★★★★★ | Text Classification 🎭 | Feb 25 |
| distilbert-base-uncased-finetuned-sst-2-english | distilbert | 0.22 | ★★★★★ | Text Classification 🎭 | Feb 25 |
| distilbert-base-multilingual-cased-sentiments-student | lxyuan | 0.22 | ★★★★★ | Text Classification 🎭 | Feb 25 |
| question_answering_v2 | Falconsai | 0.23 | ★★★★★ | Extractive QA ❔ | Feb 25 |
| distilbert-base-cased-distilled-squad | distilbert | 0.24 | ★★★★★ | Extractive QA ❔ | Feb 25 |
| gte-base | thenlper | 0.25 | ★★★★★ | Sentence Similarity 📚 | Feb 25 |
| bge-base-en-v1.5 | BAAI | 0.25 | ★★★★★ | Sentence Similarity 📚 | Feb 25 |
| electra-base-emotion | bhadresh-savani | 0.25 | ★★★★★ | Text Classification 🎭 | Feb 25 |
| minilm-uncased-squad2 | deepset | 0.25 | ★★★★★ | Extractive QA ❔ | Feb 25 |
| e5-base-v2 | intfloat | 0.25 | ★★★★★ | Sentence Similarity 📚 | Feb 25 |
| GIST-Embedding-v0 | avsolatorio | 0.26 | ★★★★★ | Sentence Similarity 📚 | Feb 25 |
| resnet-18 | microsoft | 0.30 | ★★★★★ | Image Classification 🖼️ | Feb 25 |
| gpt2-imdb-sentiment-classifier | mnoukhov | 0.32 | ★★★★★ | Text Classification 🎭 | Feb 25 |
| bert-base-multilingual-uncased-sentiment | nlptown | 0.36 | ★★★★★ | Text Classification 🎭 | Feb 25 |
| nomic-embed-text-v1 | nomic-ai | 0.38 | ★★★★★ | Sentence Similarity 📚 | Feb 25 |
| vit-tiny-patch16-224 | WinKawaks | 0.38 | ★★★★★ | Image Classification 🖼️ | Feb 25 |
| resnet-50 | microsoft | 0.39 | ★★★★★ | Image Classification 🖼️ | Feb 25 |
| mobilenet_v2_1.0_224 | 0.41 | ★★★★★ | Image Classification 🖼️ | Feb 25 | |
| convnext-tiny-224 | 0.45 | ★★★★★ | Image Classification 🖼️ | Feb 25 | |
| electra-base-squad2 | deepset | 0.47 | ★★★★★ | Extractive QA ❔ | Feb 25 |
| mobilevit-small | apple | 0.51 | ★★★★★ | Image Classification 🖼️ | Feb 25 |
| ember-v1 | llmrails | 0.58 | ★★★★★ | Sentence Similarity 📚 | Feb 25 |
| GIST-large-Embedding-v0 | avsolatorio | 0.58 | ★★★★★ | Sentence Similarity 📚 | Feb 25 |
| vit-base-patch16-224 | 0.58 | ★★★★★ | Image Classification 🖼️ | Feb 25 | |
| mxbai-embed-large-v1 | mixedbread-ai | 0.59 | ★★★★★ | Sentence Similarity 📚 | Feb 25 |
| gte-large | thenlper | 0.59 | ★★★★★ | Sentence Similarity 📚 | Feb 25 |
| UAE-Large-V1 | WhereIsAI | 0.60 | ★★★★★ | Sentence Similarity 📚 | Feb 25 |
| e5-large-v2 | intfloat | 0.60 | ★★★★★ | Sentence Similarity 📚 | Feb 25 |
| wav2vec2-base-960h | 0.62 | ★★★★★ | Automatic Speech Recognition 💬 | Feb 25 | |
| multilingual-e5-large-instruct | intfloat | 0.72 | ★★★★★ | Sentence Similarity 📚 | Feb 25 |
| beit-base-patch16-224 | microsoft | 0.74 | ★★★★★ | Image Classification 🖼️ | Feb 25 |
| vit-base-patch32-384 | 0.76 | ★★★★★ | Image Classification 🖼️ | Feb 25 | |
| swinv2-tiny-patch4-window8-256 | microsoft | 0.77 | ★★★★ | Image Classification 🖼️ | Feb 25 |
| beit-base-patch16-224-pt22k-ft22k | microsoft | 0.78 | ★★★★ | Image Classification 🖼️ | Feb 25 |
| bart_summarizer_model | KipperDev | 0.78 | ★★★★★ | Summarization 📃 | Feb 25 |
| gpt2-medium-finetuned-sst2-sentiment | michelecafagna26 | 0.80 | ★★★★★ | Text Classification 🎭 | Feb 25 |
| deberta-v3-base-squad2 | deepset | 0.81 | ★★★★★ | Extractive QA ❔ | Feb 25 |
| bit-50 | 0.83 | ★★★★ | Image Classification 🖼️ | Feb 25 | |
| swinv2-tiny-patch4-window16-256 | microsoft | 0.85 | ★★★★ | Image Classification 🖼️ | Feb 25 |
| mdeberta-v3-base-squad2 | timpal0l | 0.86 | ★★★★★ | Extractive QA ❔ | Feb 25 |
| bert-large-sst2 | Cheng98 | 0.92 | ★★★★★ | Text Classification 🎭 | Feb 25 |
| mobilebert-uncased-squad-v2 | csarron | 0.94 | ★★★★★ | Extractive QA ❔ | Feb 25 |
| udever-bloom-1b1 | izhx | 0.97 | ★★★★★ | Sentence Similarity 📚 | Feb 25 |
| convnext-large-224 | 0.97 | ★★★★ | Image Classification 🖼️ | Feb 25 | |
| vit-base-patch16-384 | 0.99 | ★★★★ | Image Classification 🖼️ | Feb 25 | |
| yolos-tiny | hustvl | 1.00 | ★★★★★ | Object Detection 🚘 | Feb 25 |
| bert-large-uncased-whole-word-masking-squad2 | deepset | 1.05 | ★★★★★ | Extractive QA ❔ | Feb 25 |
| distilbart-xsum-12-1 | sshleifer | 1.05 | ★★★★★ | Summarization 📃 | Feb 25 |
| convnext-base-384-22k-1k | 1.12 | ★★★★ | Image Classification 🖼️ | Feb 25 | |
| vit-large-patch32-384 | 1.16 | ★★★★ | Image Classification 🖼️ | Feb 25 | |
| swinv2-large-patch4-window12-192-22k | microsoft | 1.29 | ★★★ | Image Classification 🖼️ | Feb 25 |
| distilgpt2 | distilbert | 1.31 | ★★★★★ | Text Generation 💬 | Feb 25 |
| distilbart-cnn-12-6 | sshleifer | 1.48 | ★★★★★ | Summarization 📃 | Feb 25 |
| swin-base-patch4-window12-384 | microsoft | 1.53 | ★★★ | Image Classification 🖼️ | Feb 25 |
| sew-d-base-plus-400k-ft-ls100h | asapp | 1.54 | ★★★★★ | Automatic Speech Recognition 💬 | Feb 25 |
| hubert-large-ls960-ft | 1.62 | ★★★★★ | Automatic Speech Recognition 💬 | Feb 25 | |
| longformer-base-4096-finetuned-squadv2 | mrm8488 | 1.79 | ★★★★★ | Extractive QA ❔ | Feb 25 |
| efficientnet-b7 | 1.85 | ★★ | Image Classification 🖼️ | Feb 25 | |
| deberta-v3-large-squad2 | deepset | 1.88 | ★★★★★ | Extractive QA ❔ | Feb 25 |
| opt-125m | 1.94 | ★★★★★ | Text Generation 💬 | Feb 25 | |
| bart-large-cnn | 2.04 | ★★★★★ | Summarization 📃 | Feb 25 | |
| conditional-detr-resnet-50 | microsoft | 2.14 | ★★★★★ | Object Detection 🚘 | Feb 25 |
| gpt2 | openai-community | 2.15 | ★★★★★ | Text Generation 💬 | Feb 25 |
| udever-bloom-3b | izhx | 2.15 | ★★★ | Sentence Similarity 📚 | Feb 25 |
| t5-small | google-t5 | 2.22 | ★★★★★ | Text Classification 🎭 | Feb 25 |
| mit-b5 | nvidia | 2.49 | ★ | Image Classification 🖼️ | Feb 25 |
| detr-resnet-101 | 2.53 | ★★★★ | Object Detection 🚘 | Feb 25 | |
| vision-perceiver-conv | deepmind | 2.64 | ★ | Image Classification 🖼️ | Feb 25 |
| openai-gpt | openai-community | 2.70 | ★★★★★ | Text Generation 💬 | Feb 25 |
| yolos-small | hustvl | 3.28 | ★★★★ | Object Detection 🚘 | Feb 25 |
| detr-resnet-50-dc5 | 3.29 | ★★★★ | Object Detection 🚘 | Feb 25 | |
| bigbird-pegasus-large-arxiv | 3.52 | ★★★★★ | Summarization 📃 | Feb 25 | |
| gpt2-medium | openai-community | 3.96 | ★★★★★ | Text Generation 💬 | Feb 25 |
| longformer-large-4096-finetuned-triviaqa | allenai | 3.97 | ★★★★★ | Extractive QA ❔ | Feb 25 |
| t5-base | google-t5 | 4.13 | ★★★★★ | Text Classification 🎭 | Feb 25 |
| mms-1b-all | 4.28 | ★★★★★ | Automatic Speech Recognition 💬 | Feb 25 | |
| led-large-book-summary | pszemraj | 4.29 | ★★★★★ | Summarization 📃 | Feb 25 |
| udever-bloom-7b1 | izhx | 4.70 | ★ | Sentence Similarity 📚 | Feb 25 |
| GritLM-7B | GritLM | 5.03 | ★ | Sentence Similarity 📚 | Feb 25 |
| SFR-Embedding-Mistral | Salesforce | 5.22 | ★ | Sentence Similarity 📚 | Feb 25 |
| OLMo-1B-hf | allenai | 5.59 | ★★★★★ | Text Generation 💬 | Feb 25 |
| gpt2-large | openai-community | 6.12 | ★★★★★ | Text Generation 💬 | Feb 25 |
| phi-1_5 | microsoft | 6.29 | ★★★★★ | Text Generation 💬 | Feb 25 |
| phi-1 | microsoft | 6.39 | ★★★★★ | Text Generation 💬 | Feb 25 |
| SmolLM-135M | HuggingFaceTB | 7.28 | ★★★★ | Text Generation 💬 | Feb 25 |
| SmolLM-1.7B | HuggingFaceTB | 7.64 | ★★★★ | Text Generation 💬 | Feb 25 |
| SmolLM2-1.7B-Instruct | HuggingFaceTB | 7.71 | ★★★★ | Text Generation 💬 | Feb 25 |
| t5-large | google-t5 | 7.92 | ★★★★ | Text Classification 🎭 | Feb 25 |
| yolos-base | hustvl | 7.98 | ★ | Object Detection 🚘 | Feb 25 |
| opt-2.7b | 8.06 | ★★★★ | Text Generation 💬 | Feb 25 | |
| SmolLM-360M | HuggingFaceTB | 8.10 | ★★★★ | Text Generation 💬 | Feb 25 |
| gemma-2b | 8.68 | ★★★★ | Text Generation 💬 | Feb 25 | |
| gpt2-xl | openai-community | 8.92 | ★★★★ | Text Generation 💬 | Feb 25 |
| gemma-3-270m | 9.29 | ★★★★★ | Text Generation 💬 | Feb 25 | |
| phi-2 | microsoft | 9.86 | ★★★★ | Text Generation 💬 | Feb 25 |
| t5-3b | google-t5 | 11.68 | ★★★ | Text Classification 🎭 | Feb 25 |
| gemma-3-1b-pt | 12.33 | ★★★★★ | Text Generation 💬 | Feb 25 | |
| whisper-tiny | openai | 13.44 | ★★★★★ | Automatic Speech Recognition 💬 | Feb 25 |
| distil-small.en | distil-whisper | 15.57 | ★★★★★ | Automatic Speech Recognition 💬 | Feb 25 |
| falcon-7b | tiiuae | 15.88 | ★★★ | Text Generation 💬 | Feb 25 |
| Llama-2-7b-hf | meta-llama | 16.44 | ★★★ | Text Generation 💬 | Feb 25 |
| whisper-base | openai | 16.55 | ★★★★★ | Automatic Speech Recognition 💬 | Feb 25 |
| granite-4.0-micro | ibm-granite | 16.63 | ★★★★★ | Text Generation 💬 | Feb 25 |
| OLMo-1.7-7B-hf | allenai | 16.83 | ★★★ | Text Generation 💬 | Feb 25 |
| distil-medium.en | distil-whisper | 16.97 | ★★★★★ | Automatic Speech Recognition 💬 | Feb 25 |
| s2t-small-librispeech-asr | 17.20 | ★★★★★ | Automatic Speech Recognition 💬 | Feb 25 | |
| Meta-Llama-3-8B | meta-llama | 17.33 | ★★★ | Text Generation 💬 | Feb 25 |
| Meta-Llama-3.1-8B | meta-llama | 17.38 | ★★★ | Text Generation 💬 | Feb 25 |
| aya-23-8B | CohereForAI | 17.82 | ★★ | Text Generation 💬 | Feb 25 |
| gemma-2-2b-it | 18.64 | ★★ | Text Generation 💬 | Feb 25 | |
| gemma-7b | 18.90 | ★★ | Text Generation 💬 | Feb 25 | |
| Starling-LM-7B-alpha | berkeley-nest | 19.01 | ★★ | Text Generation 💬 | Feb 25 |
| Mistral-7B-v0.1 | mistralai | 19.13 | ★★ | Text Generation 💬 | Feb 25 |
| xLAM-7b-r | Salesforce | 19.17 | ★★ | Text Generation 💬 | Feb 25 |
| Ministral-8B-Instruct-2410 | mistralai | 20.88 | ★★ | Text Generation 💬 | Feb 25 |
| distil-large-v2 | distil-whisper | 21.52 | ★★★★ | Automatic Speech Recognition 💬 | Feb 25 |
| Apertus-8B-Instruct-2509 | swiss-ai | 21.74 | ★★★★ | Text Generation 💬 | Feb 25 |
| Yi-1.5-9B | 01-ai | 23.00 | ★★ | Text Generation 💬 | Feb 25 |
| Mistral-Nemo-Instruct-2407 | mistralai | 23.19 | ★★★★ | Text Generation 💬 | Feb 25 |
| Llama-2-13b-hf | meta-llama | 26.10 | ★ | Text Generation 💬 | Feb 25 |
| Qwen2.5-Coder-14B | Qwen | 27.12 | ★★★★ | Text Generation 💬 | Feb 25 |
| t5-11b | google-t5 | 27.79 | ★ | Text Classification 🎭 | Feb 25 |
| phi-4 | microsoft | 28.74 | ★ | Text Generation 💬 | Feb 25 |
| whisper-small | openai | 29.97 | ★★★★ | Automatic Speech Recognition 💬 | Feb 25 |
| pix2struct-base | 45.75 | ★★★★★ | Image Captioning 📝 | Feb 25 | |
| blip-image-captioning-base | Salesforce | 47.12 | ★★★★★ | Image Captioning 📝 | Feb 25 |
| blip-image-captioning-large | Salesforce | 48.51 | ★★★★★ | Image Captioning 📝 | Feb 25 |
| gemma-3-27b-pt | 50.41 | ★★ | Text Generation 💬 | Feb 25 | |
| QwQ-32B-Preview | Qwen | 54.56 | ★★★★★ | Text Generation 💬 | Feb 25 |
| Yi-1.5-34B | 01-ai | 56.08 | ★★★★★ | Text Generation 💬 | Feb 25 |
| whisper-medium | openai | 61.42 | ★★ | Automatic Speech Recognition 💬 | Feb 25 |
| git-base | microsoft | 74.71 | ★★★★ | Image Captioning 📝 | Feb 25 |
| granite-4.0-h-tiny | ibm-granite | 81.82 | ★ | Text Generation 💬 | Feb 25 |
| whisper-large-v3 | openai | 86.68 | ★ | Automatic Speech Recognition 💬 | Feb 25 |
| blip2-opt-2.7b | Salesforce | 101.34 | ★★★ | Image Captioning 📝 | Feb 25 |
| blip2-flan-t5-xl | Salesforce | 102.15 | ★★★ | Image Captioning 📝 | Feb 25 |
| vlrm-blip2-opt-2.7b | sashakunitsyn | 103.35 | ★★★ | Image Captioning 📝 | Feb 25 |
| git-large-coco | microsoft | 170.32 | ★ | Image Captioning 📝 | Feb 25 |
| blip2-opt-6.7b | Salesforce | 174.96 | ★ | Image Captioning 📝 | Feb 25 |
| blip2-flan-t5-xxl | Salesforce | 179.93 | ★ | Image Captioning 📝 | Feb 25 |
| mitsua-diffusion-one | Mitsua | 186.81 | ★★★★★ | Image Generation 📷 | Feb 25 |
| Counterfeit-V2.5 | gsdf | 187.37 | ★★★★★ | Image Generation 📷 | Feb 25 |
| sd-turbo | stabilityai | 189.86 | ★★★★★ | Image Generation 📷 | Feb 25 |
| openjourney | prompthero | 196.88 | ★★★★★ | Image Generation 📷 | Feb 25 |
| openjourney-v4 | prompthero | 203.37 | ★★★★★ | Image Generation 📷 | Feb 25 |
| GLM-4.5-Air | zai-org | 220.55 | ★★★★★ | Text Generation 💬 | Feb 25 |
| Qwen3-Next-80B-A3B-Instruct | Qwen | 266.73 | ★★★★★ | Reasoning 🧠 | Feb 25 |
| Qwen3-Next-80B-A3B-Thinking | Qwen | 276.67 | ★★★★★ | Reasoning 🧠 | Feb 25 |
| Qwen3-235B-A22B | Qwen | 300.53 | ★★★★★ | Reasoning 🧠 | Feb 25 |
| LCM_Dreamshaper_v7 | SimianLuo | 322.04 | ★★★★★ | Image Generation 📷 | Feb 25 |
| sdxl-turbo | stabilityai | 385.57 | ★★★★★ | Image Generation 📷 | Feb 25 |
| stable-diffusion-2-1 | stabilityai | 534.47 | ★★★★ | Image Generation 📷 | Feb 25 |
| Qwen3-0.6B | Qwen | 546.99 | ★★★★★ | Reasoning 🧠 | Feb 25 |
| dreamlike-photoreal-2.0 | dreamlike-art | 581.33 | ★★★★ | Image Generation 📷 | Feb 25 |
| epiCPhotoGasm | Yntec | 586.57 | ★★★★ | Image Generation 📷 | Feb 25 |
| Mixtral-8x7B-v0.1 | mistralai | 615.39 | ★ | Text Generation 💬 | Feb 25 |
| Apertus-70B-Instruct-2509 | swiss-ai | 697.52 | ★★★ | Text Generation 💬 | Feb 25 |
| gpt-oss-20b low | openai | 1,135.59 | ★★★★★ | Reasoning 🧠 | Feb 25 |
| stable-cascade | stabilityai | 1,214.36 | ★★ | Image Generation 📷 | Feb 25 |
| GLM-4.5 | zai-org | 1,225.42 | ★ | Text Generation 💬 | Feb 25 |
| GLM-4.6 | zai-org | 1,273.55 | ★ | Text Generation 💬 | Feb 25 |
| Hunyuan-1.8B-Instruct | tencent | 1,478.96 | ★★★★★ | Reasoning 🧠 | Feb 25 |
| stable-diffusion-xl-base-1.0 | stabilityai | 1,639.85 | ★ | Image Generation 📷 | Feb 25 |
| Llama-2-70b-hf | meta-llama | 1,648.51 | ★★★★★ | Text Generation 💬 | Feb 25 |
| Meta-Llama-3-70B | meta-llama | 1,719.66 | ★★★★★ | Text Generation 💬 | Feb 25 |
| TextBase-70B | crm-sf | 1,721.56 | ★★★★★ | Text Generation 💬 | Feb 25 |
| Qwen2.5-72B-Instruct | Qwen | 1,869.55 | ★★★★★ | Text Generation 💬 | Feb 25 |
| c4ai-command-r-plus | CohereForAI | 3,426.38 | ★ | Text Generation 💬 | Feb 25 |
| gpt-oss-20b medium | openai | 4,122.11 | ★★★★★ | Reasoning 🧠 | Feb 25 |
| DeepSeek-R1-Distill-Qwen-32B | deepseek-ai | 4,789.83 | ★★★ | Reasoning 🧠 | Feb 25 |
| gpt-oss-120b low | openai | 5,312.93 | ★★ | Reasoning 🧠 | Feb 25 |
| gpt-oss-20b high | openai | 5,512.01 | ★★★★ | Reasoning 🧠 | Feb 25 |
| gpt-oss-120b medium | openai | 6,586.30 | ★★ | Reasoning 🧠 | Feb 25 |
| DeepSeek-R1-Distill-Llama-70B | deepseek-ai | 7,626.53 | ★ | Reasoning 🧠 | Feb 25 |
| gpt-oss-120b high | openai | 8,503.62 | ★ | Reasoning 🧠 | Feb 25 |
| Phi-4-reasoning-plus | microsoft | 9,461.61 | ★★ | Reasoning 🧠 | Feb 25 |
| Qwen3-30B-A3B | Qwen | 11,130.06 | ★★★ | Reasoning 🧠 | Feb 25 |
| gemma-3-4b-pt | 11,811.18 | ★ | Reasoning 🧠 | Feb 25 | |
| SmolLM3-3B | HuggingFaceTB | 12,791.22 | ★ | Reasoning 🧠 | Feb 25 |
| EXAONE-4.0-32B | LGAI-EXAONE | 18,992.36 | ★ | Reasoning 🧠 | Feb 25 |
Last updated: December 2025