| Task | Model | Params (billions) | Notes |
|---|---|---|---|
| Image Object Detection | DETR-DC5 R101 | 0.607 | |
| Image Masking | Segment Anything + ViT Huge | 0.641 | |
| Image Depth Map Creation | Depth Anything v2 Huge | 1.3 | not released yet, only Small to Large |
| Caption to Image | Stable Diffusion XL | 2.6 | |
| Caption to Video | Open-Sora | ||
| Image to Video | Stable Video Diffusion XT | ||
| Caption to Sound | |||
| Image to 3D | InstantMesh | taken from 3D Arena / maybe SV3D | |
| Coding LLM (Instruct) | DeepSeek Coder v2 Instruct | 236 | |
| Code LLM (Completion / Filling holes) | DeepSeek Coder v2 Base | 236 | |
| General LLM (Instruct) | Llama 3 | 70.6 | |
| General LLM (Completion / Filling holes) | Roberta Large | 0.355 | |
| Text to Speech 🇺🇸 | Parler | 0.6 | |
| Speech to Text 🇺🇸 | Whisper Large v3 | 1.54 | |
| Text to Speech 🇩🇪 | |||
| Speech to Text 🇩🇪 | |||
| Speech to Language | Lang ID | ||
| Speaking animation for portraits | Hallo |
Last active
June 20, 2024 00:15
-
-
Save Jaid/5ddee6cf20145da1b1b8745ef23b59e1 to your computer and use it in GitHub Desktop.
Current best open-source AI models
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment