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 |
|---|---|---|---|---|---|---|
| whisper-medium openai | Automatic speech recognition | — | — | 0.8B | 354.9K | Apache 2.0 |
| llama-nemotron-embed-1b-v2 nvidia | Feature extraction | ~2.8 GB | €0.04/h RTX 3060 | 1.2B | 347.1K | Other |
| Wan2.1-T2V-1.3B-Diffusers Wan-AI | Text to video | — | — | 1.4B | 346.7K | Apache 2.0 |
| wav2vec2-large-xls-r-300m-sinhala-low-LR-part1 SpideyDLK | Automatic speech recognition | — | — | 0.3B | 333.4K | — |
| Qwen3-ASR-1.7B-hf Qwen | Automatic speech recognition | — | — | 2B | 331.9K | Apache 2.0 |
| Z-Image-Turbo unsloth | Text to image | — | — | 6.2B | 328.8K | Apache 2.0 |
| Qwen-Image Qwen | Text to image | — | — | 20.4B | 325.2K | Apache 2.0 |
| wav2vec2-large-xlsr-marathi sumedh | Automatic speech recognition | — | — | 0.3B | 322.5K | Apache 2.0 |
| Solon-embeddings-large-0.1 OrdalieTech | Feature extraction | ~1.3 GB | €0.04/h RTX 3060 | 0.6B | 306.7K | MIT |
| Krea-2-Turbo unsloth | Text to image | — | — | 12.8B | 301.5K | Other |
| LTX-2 Lightricks | Image to video | — | — | 18.9B | 295.4K | Other |
| animagine-xl-3.1 cagliostrolab | Text to image | — | — | 2.6B | 286.4K | OpenRAIL++ |
| controlnet-openpose-sdxl-1.0 xinsir | Text to image | — | — | 1.3B | 284.9K | Apache 2.0 |
| NV-Embed-v2 nvidia | Feature extraction | ~18 GB | €0.17/h RTX 3090 | 7.9B | 278.3K | CC BY-NC 4.0 |
| FLUX.1-schnell black-forest-labs | Text to image | — | — | 11.9B | 278K | Apache 2.0 |
| KURE-v1 nlpai-lab | Feature extraction | ~1.3 GB | €0.04/h RTX 3060 | 0.6B | 269K | MIT |
| jina-embeddings-v3-hf jinaai | Feature extraction | ~1.3 GB | €0.04/h RTX 3060 | 0.6B | 238.8K | CC BY-NC 4.0 |
| cohere-transcribe-arabic-07-2026 CohereLabs | Automatic speech recognition | — | — | 2.1B | 236.3K | Apache 2.0 |
| MoLFormer-XL-both-10pct ibm-research | Feature extraction | ~0.1 GB | €0.04/h RTX 3060 | 0B | 234.8K | Apache 2.0 |
| canary-1b-v2 nvidia | Automatic speech recognition | — | — | 1B | 233.8K | CC BY 4.0 |
| chatterbox-turbo-4bit mlx-community | Text to speech | — | — | 0.6B | 231.4K | Apache 2.0 |
| RealVisXL_V4.0 SG161222 | Text to image | — | — | 2.6B | 223K | OpenRAIL++ |
| harrier-oss-v1-0.6b microsoft | Feature extraction | ~1.3 GB | €0.04/h RTX 3060 | 0.6B | 221.7K | MIT |
| MOSS-Audio-Tokenizer-Nano OpenMOSS-Team | Feature extraction | ~0 GB | €0.04/h RTX 3060 | 0B | 216.9K | Apache 2.0 |
| bart-base facebook | Feature extraction | ~0.3 GB | €0.04/h RTX 3060 | 0.1B | 216.1K | Apache 2.0 |