Malayalam LLM fine-tuning and inference
10 repos
Large language model implementations and adaptations optimized for Malayalam language processing, built on Gemma 2 and related architectures. These repositories focus on making instruction-tuned and multimodal variants compatible with standard text generation inference frameworks, leveraging safetensors for efficient model distribution. The cluster covers both dense model variants and quantized GGUF formats for resource-constrained deployments.
Python · 4
endpoints_compatible ·5
gemma2 ·4
transformers ·4
text-generation ·4
text-generation-inference ·4
safetensors ·4
gguf ·1
paligemma ·0
