Compare LLM Models
Select 2-5 models to compare side by side.
Select models to compare
0 / 5 selected
Top downloads
Select 2-5 models
| Attribute |
HuggingFaceTB
|
|---|---|
| Company | HuggingFaceTB |
| Release Date | 2024-11-26 |
| Parameters (B) | 2.2 |
| Architecture | Idefics3 Vision-Language Transformer |
| Context Window | 16384 |
| Input Modalities | text, image |
| Open Source | Yes |
| License | apache-2.0 |
| Score | 0 |
| VRAM (GB) | 5.0 |
| Compute | Consumer GPU with ~5 GB VRAM for BF16 single-image inference; CUDA recommended with Flash Attention 2; QLoRA fine-tuning feasible on ~16 GB GPUs (e.g., NVIDIA L4) |
| Benchmark: MMMU | 38.8 |
| Pricing (Input) | — |
| Pricing (Output) | — |