kkt94/Llama-V-JEPA-LM-Qwen3-1.7B
Video-Text-to-Text • 2B • Updated • 219 • 11
Video-language models that read V-JEPA dynamics (1B-8B, Qwen3 and Llama). Technical report and code: github.com/kkt94/Llama-V-JEPA-LM
Video question answering with V-JEPA 2.1 + Llama 3.1 8B
Note Live demo (Llama3.1-8B, ZeroGPU)
Note Paper (arXiv:2611.17831)