Vocaela-500M-GGUF

This repo contains GGUF-format weights for Vocaela-500M, containing:

• Language model (LLM): Q8_0
• Vision encoder (mmproj): Q8_0

Note: So far LlamaCpp has problem on rendering chat template correctly. To workaround it, we apply chat template (e.g., using python / node.js etc.) before calling llama-server endpoint. For examples of how to use it, please refer to repo vocaela-500m-demo