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 |
|---|---|---|---|---|---|---|
| animagine-xl-4.0 cagliostrolab | Text to image | — | — | 2.6B | 1M | OpenRAIL++ |
| Qwen2-VL-2B-Instruct Qwen | Image text to text | ~4.9 GB | €0.04/h RTX 3060 | 2.2B | 1M | Apache 2.0 |
| Ternary-Bonsai-27B-mlx-2bit prism-ml | Text generation | ~61 GB | €2.09/h RTX PRO 6000 | 27.4B | 1M | Apache 2.0 |
| jina-embeddings-v5-omni-small jinaai | Feature extraction | ~3.6 GB | €0.04/h RTX 3060 | 1.6B | 1M | CC BY-NC 4.0 |
| NVIDIA-Nemotron-3.5-Lightning-30B-A3B-NVFP4 nvidia | Text generation | ~39 GB | €0.62/h RTX 6000 Ada | 32.9B | 1M | Other |
| UAE-Large-V1 WhereIsAI | Feature extraction | ~0.7 GB | €0.04/h RTX 3060 | 0.3B | 1M | MIT |
| gemma-3-4b-it google | Image text to text | ~9.6 GB | €0.09/h RTX 3080 | 4.3B | 1M | Gemma |
| Bonsai-27B-mlx-1bit prism-ml | Text generation | ~3.9 GB | €0.04/h RTX 3060 | 1.7B | 1M | Apache 2.0 |
| wav2vec2-xls-r-300m-hebrew imvladikon | Automatic speech recognition | — | — | 0.3B | 1M | — |
| VoxCPM2 openbmb | Text to speech | — | — | 2.3B | 987K | Apache 2.0 |
| surya-ocr-2 datalab-to | Image text to text | ~1.5 GB | €0.04/h RTX 3060 | 0.7B | 975.4K | OpenRAIL |
| Qwen2.5-Coder-32B-Instruct-AWQ Qwen | Text generation | ~73 GB | €2.09/h RTX PRO 6000 | 32.8B | 971.5K | Apache 2.0 |
| Qwen3.8-27B-NVFP4 RadixArk | Image text to text | ~23 GB | €0.17/h RTX 3090 | 18.6B | 969.2K | Apache 2.0 |
| InternVL2-1B OpenGVLab | Image text to text | ~2.1 GB | €0.04/h RTX 3060 | 0.9B | 963.9K | MIT |
| nb-wav2vec2-1b-bokmaal-v2 NbAiLab | Automatic speech recognition | — | — | 1B | 946.7K | Apache 2.0 |
| TinyLlama-1.1B-Chat-v1.0 TinyLlama | Text generation | ~2.5 GB | €0.04/h RTX 3060 | 1.1B | 943.9K | Apache 2.0 |
| Wav2Vec2-large-xlsr-hindi theainerd | Automatic speech recognition | — | — | 0.3B | 932.1K | — |
| Qwen2.5-32B-Instruct-AWQ Qwen | Text generation | ~73 GB | €2.09/h RTX PRO 6000 | 32.8B | 922.6K | Apache 2.0 |
| gpt2-large openai-community | Text generation | ~1.8 GB | €0.04/h RTX 3060 | 0.8B | 907.7K | MIT |
| Qwen3.6-35B-A3B-NVFP4 RedHatAI | Image text to text | ~41 GB | €0.62/h RTX 6000 Ada | 34.7B | 906K | Apache 2.0 |
| Qwen3.5-122B-A10B Qwen | Image text to text | ~280 GB | — | 125B | 904.2K | Apache 2.0 |
| dreamshaper-7 Lykon | Text to image | — | — | 0.9B | 902.2K | creativeml-openrail-m |
| Qwen3-30B-A3B-Instruct-2507 Qwen | Text generation | ~68 GB | €2.09/h RTX PRO 6000 | 30.5B | 895.7K | Apache 2.0 |
| wav2vec2-xls-r-300m-ftspeech saattrupdan | Automatic speech recognition | — | — | 0.3B | 888.2K | Other |
| GLM-5.1 zai-org | Text generation | ~1685 GB | — | 754B | 878.9K | MIT |