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 |
|---|---|---|---|---|---|---|
| Florence-2-base microsoft | Image text to text | ~0.5 GB | €0.04/h RTX 3060 | 0.2B | 11.7M | MIT |
| Qwen3-Embedding-0.6B Qwen | Feature extraction | ~1.3 GB | €0.04/h RTX 3060 | 0.6B | 9.1M | Apache 2.0 |
| bge-base-en-v1.5 BAAI | Feature extraction | ~0.2 GB | €0.04/h RTX 3060 | 0.1B | 8.3M | MIT |
| bge-large-en-v1.5 BAAI | Feature extraction | ~0.7 GB | €0.04/h RTX 3060 | 0.3B | 8.2M | MIT |
| DeepSeek-V3.2 deepseek-ai | Text generation | ~770 GB | — | 685B | 7.6M | MIT |
| multilingual-e5-large intfloat | Feature extraction | ~1.3 GB | €0.04/h RTX 3060 | 0.6B | 7.1M | MIT |
| granite-embedding-small-english-r2 ibm-granite | Feature extraction | ~0.1 GB | €0.04/h RTX 3060 | 0B | 6.9M | Apache 2.0 |
| Qwen3-8B Qwen | Text generation | ~18 GB | €0.17/h RTX 3090 | 8.2B | 6.8M | Apache 2.0 |
| GLM-5.3-Flash zai-org | Image text to text | ~367 GB | — | 321B | 6.4M | MIT |
| Qwen2.5-7B-Instruct Qwen | Text generation | ~17 GB | €0.17/h RTX 3090 | 7.6B | 6M | Apache 2.0 |
| Qwen2.5-VL-7B-Instruct Qwen | Image text to text | ~19 GB | €0.17/h RTX 3090 | 8.3B | 6M | Apache 2.0 |
| gemma-4-26B-A4B-it google | Image text to text | ~58 GB | €2.09/h RTX PRO 6000 | 25.8B | 5.5M | Apache 2.0 |
| Unlimited-OCR baidu | Image text to text | ~7.5 GB | €0.04/h RTX 3060 | 3.3B | 5.4M | MIT |
| Qwen3.8-27B-FP8 Qwen | Image text to text | ~34 GB | €0.62/h RTX 6000 Ada | 27.8B | 5.3M | Apache 2.0 |
| Qwen3.5-4B Qwen | Image text to text | ~10 GB | €0.13/h RTX 5070 | 4.7B | 5.3M | Apache 2.0 |
| Qwen2.5-1.5B-Instruct Qwen | Text generation | ~3.5 GB | €0.04/h RTX 3060 | 1.5B | 5M | Apache 2.0 |
| dolphin-2.9.1-yi-1.5-34b dphn | Text generation | ~77 GB | €2.09/h RTX PRO 6000 | 34.4B | 4.8M | Apache 2.0 |
| whisper-base openai | Automatic speech recognition | — | — | 0.1B | 4.6M | Apache 2.0 |
| Qwen2.5-0.5B-Instruct Qwen | Text generation | ~1.1 GB | €0.04/h RTX 3060 | 0.5B | 4.5M | Apache 2.0 |
| whisper-small openai | Automatic speech recognition | — | — | 0.2B | 4.4M | Apache 2.0 |
| Qwen3.6-35B-A3B Qwen | Image text to text | ~80 GB | €2.09/h RTX PRO 6000 | 36B | 4.2M | Apache 2.0 |
| Qwen3-4B Qwen | Text generation | ~9 GB | €0.09/h RTX 3080 | 4B | 4M | Apache 2.0 |
| Qwen3.6-35B-A3B-NVFP4 nvidia | Text generation | ~24 GB | €0.17/h RTX 3090 | 19.5B | 4M | Apache 2.0 |
| Qwen3.6-35B-A3B-FP8 Qwen | Image text to text | ~42 GB | €0.62/h RTX 6000 Ada | 36B | 4M | Apache 2.0 |
| Qwen3-VL-4B-Instruct Qwen | Image text to text | ~9.9 GB | €0.09/h RTX 3080 | 4.4B | 3.7M | Apache 2.0 |