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 |
|---|---|---|---|---|---|---|
| Qwen3.8 27B Qwen | Image text to text | 67 GB | €2.09/h RTX PRO 6000 | 27.8B | 7.5M | Apache 2.0 |
| Qwen3-VL 8B Qwen | Image text to text | 20 GB | €0.17/h RTX 3090 | 8.8B | 7.1M | Apache 2.0 |
| Gemma 4 31B google | Image text to text | 75 GB | €2.09/h RTX PRO 6000 | 31.3B | 7.1M | Apache 2.0 |
| Qwen3.5 9B Qwen | Image text to text | 22 GB | €0.17/h RTX 3090 | 9.7B | 6.4M | Apache 2.0 |
| gpt-oss-20b openai | Text generation | 16 GB | €0.17/h RTX 3090 | 20.9B | 5.8M | Apache 2.0 |
| Whisper large-v3-turbo openai | Automatic speech recognition | 6 GB | €0.04/h RTX 3060 | 0.8B | 5.2M | MIT |
| Stable Diffusion XL 1.0 stabilityai | Text to image | 7.8 GB | €0.04/h RTX 3060 | 2.6B | 4.6M | OpenRAIL++ |
| Llama 3.1 8B Instruct meta-llama | Text generation | 16 GB | €0.17/h RTX 3090 | 8B | 4.5M | Llama 3.1 |
| Whisper large-v3 openai | Automatic speech recognition | 10 GB | €0.09/h RTX 3080 | 1.5B | 3.5M | Apache 2.0 |
| Qwen3.8 Flash Next Qwen | Image text to text | 423 GB | — | 180B | 3.5M | Other |
| gpt-oss-120b openai | Text generation | 80 GB | €2.09/h RTX PRO 6000 | 117B | 2.4M | Apache 2.0 |
| Qwen3-Embedding 8B Qwen | Feature extraction | 17 GB | €0.17/h RTX 3090 | 7.6B | 2.4M | Apache 2.0 |
| GLM-5.3 zai-org | Text generation | 893 GB | — | 753B | 2M | Other |
| DeepSeek-OCR deepseek-ai | Image text to text | 7 GB | €0.04/h RTX 3060 | 3.3B | 2M | MIT |
| Qwen3-TTS 1.7B CustomVoice Qwen | Text to speech | 16 GB | €0.17/h RTX 3090 | 1.9B | 1.4M | Apache 2.0 |
| Z-Image Turbo Tongyi-MAI | Text to image | 16 GB | €0.17/h RTX 3090 | 6.2B | 715K | Apache 2.0 |
| Parakeet TDT 0.6B v3 nvidia | Automatic speech recognition | 2.8 GB | €0.04/h RTX 3060 | 0.6B | 539.9K | CC BY 4.0 |
| FLUX.1 [dev] black-forest-labs | Text to image | 24 GB | €0.17/h RTX 3090 | 11.9B | 495.6K | Other |
| EmbeddingGemma 2 google | Feature extraction | 1.7 GB | €0.04/h RTX 3060 | 0.7B | 475.4K | Apache 2.0 |
| Unlimited-OCR 2 baidu | Image text to text | 7.5 GB | €0.04/h RTX 3060 | 3.3B | 428.1K | MIT |
| DeepSeek V4.1 Flash (1017) deepseek-ai | Image text to text | 614 GB | — | 763B | 312.3K | MIT |
| Qwen-Image 2.1 Qwen | Text to image | 40 GB | €0.62/h RTX 6000 Ada | 7.1B | 253.2K | Other |
| Muse Glimmer 30B meta-models | Image text to text | 64 GB | €2.09/h RTX PRO 6000 | 29.8B | 179.1K | Apache 2.0 |
| LFM3 1.7B LiquidAI | Text generation | 3.9 GB | €0.04/h RTX 3060 | 1.7B | 78.8K | Other |
| Nemotron 3.5 Transcribe 30B nvidia | Automatic speech recognition | 80 GB | €2.09/h RTX PRO 6000 | 33.7B | 70K | Other |