What does your model need?
Find an open model, see the GPU memory it takes to run or fine-tune, and the cheapest GPU that fits.
652 models. Memory marked ~ is an estimate from the model's size. Prices are full price for one GPU in stock, without VAT.
| Model | Task | GPU memory | Cheapest GPU that fits | Parameters | Downloads | License |
|---|---|---|---|---|---|---|
| FRIDA ai-forever | Feature extraction | ~1.8 GB | €0.04/h RTX 3060 | 0.8B | 698.2K | MIT |
| noobai-XL-1.1 Laxhar | Text to image | — | — | 2.6B | 661.9K | Other |
| whisper-tiny openai | Automatic speech recognition | — | — | 0B | 650.2K | Apache 2.0 |
| Qwen3-ASR-0.6B Qwen | Automatic speech recognition | — | — | 0.9B | 645K | Apache 2.0 |
| sdxl-turbo stabilityai | Text to image | — | — | 2.6B | 626K | Other |
| Qwen3-TTS-12Hz-1.7B-VoiceDesign Qwen | Text to speech | — | — | 1.9B | 608.1K | Apache 2.0 |
| wav2vec2-large-xlsr-mvc-swahili eddiegulay | Automatic speech recognition | — | — | 0.3B | 599.9K | Apache 2.0 |
| Cosmos-Reason2-2B nvidia | Image text to text | ~5.5 GB | €0.04/h RTX 3060 | 2.4B | 570.7K | Other |
| nemotron-3.5-asr-streaming-0.6b nvidia | Automatic speech recognition | — | — | 0.6B | 560.1K | Other |
| Qwen-Image-2.1 unsloth | Text to image | — | — | 7.1B | 553.3K | Other |
| Qwen3.8-27B-OBLITERATED OBLITERATUS | Text generation | ~62 GB | €2.09/h RTX PRO 6000 | 27.8B | 533.3K | Apache 2.0 |
| Qwen3-ASR-0.6B-hf Qwen | Automatic speech recognition | — | — | 0.8B | 523.8K | Apache 2.0 |
| deepset-mxbai-embed-de-large-v1 mixedbread-ai | Feature extraction | ~1.1 GB | €0.04/h RTX 3060 | 0.5B | 504.3K | Apache 2.0 |
| stable-diffusion-v1-4 CompVis | Text to image | — | — | 0.9B | 482.9K | creativeml-openrail-m |
| Llama-3.1-8B meta-llama | Text generation | ~18 GB | €0.17/h RTX 3090 | 8B | 475.2K | Llama 3.1 |
| bge-small-en-v1.5 unsloth | Feature extraction | ~0.1 GB | €0.04/h RTX 3060 | 0B | 470.3K | MIT |
| wav2vec2-large-xlsr-kazakh aismlv | Automatic speech recognition | — | — | 0.3B | 465.8K | Apache 2.0 |
| sd-turbo stabilityai | Text to image | — | — | 0.9B | 453.5K | — |
| Qwen3-ForcedAligner-0.6B Qwen | Automatic speech recognition | — | — | 0.9B | 428.8K | Apache 2.0 |
| Qwen-Image-2.1-Turbo-FP8 unsloth | Text to image | — | — | 7.1B | 427.4K | Other |
| llama2-embedding-1b-8k mesolitica | Feature extraction | ~2.1 GB | €0.04/h RTX 3060 | 0.9B | 416.4K | — |
| Meta-Llama-3-8B-Instruct meta-llama | Text generation | ~18 GB | €0.17/h RTX 3090 | 8B | 385.5K | Llama 3 |
| humanizer jialinyyzz | Text generation | ~27 GB | €0.62/h RTX 6000 Ada | 12B | 379.7K | Apache 2.0 |
| uzbek-stt-3 sarahai | Automatic speech recognition | — | — | 0.3B | 358K | Apache 2.0 |
| jina-embeddings-v2-base-en jinaai | Feature extraction | ~0.3 GB | €0.04/h RTX 3060 | 0.1B | 356.9K | Apache 2.0 |