Mesolitica's repositories
NLP-Models-Tensorflow
Gathers machine learning and Tensorflow deep learning models for NLP problems, 1.13 < Tensorflow < 2.0
malaysian-dataset
We gather Malaysian dataset! https://malaysian-dataset.readthedocs.io/
malaya-speech
Speech Toolkit for Malaysian language, https://malaya-speech.readthedocs.io/
llm-embedding
Finetune Malaysian LLM for Malaysian context embedding task.
multimodal-LLM
Multi-Modal Language Modeling with Image, Audio and Text Integration, included multi-images and multi-audio in a single multiturn.
malaysian-stt-benchmarks
Benchmarking Malaysian Speech-to-Text models.
llm-benchmarks
Benchmarking LLM for Malay tasks.
whisper-static-cache
π€ Transformers: State-of-the-art Machine Learning for Pytorch, TensorFlow, and JAX.
streaming-playwright-agent
Super simple website application streaming Playwright agent using websocket that support multi-users including queue, 100% no LangChain, 100% no LlamaIndex, 100% use available models from HuggingFace.
unsloth-multigpu
Finetune Llama 3, Mistral, Phi & Gemma LLMs 2-5x faster with 80% less memory
embedding-benchmarks
Benchmarking Embedding models for Malaysian context.
huggingface-openai-api
OpenAI compatibility using FastAPI and Vanilla bitsandbytes HuggingFace Transformers, because you just don't care about speed.
malaya-boilerplate
Tensorflow freeze graph optimization and boilerplates to share among Malaya projects.
malaysian-llmops
End-to-End Malaysian LLMOps.
research-paper
Preprint under Mesolitica.
transformers
π€ Transformers: State-of-the-art Machine Learning for Pytorch, TensorFlow, and JAX.
jgscm
Jupyter support for Google Cloud Storage
nous-psyche-nft-v2-subgraph
Nous Psyche The Graph
t5-sdpa
π€ Transformers: State-of-the-art Machine Learning for Pytorch, TensorFlow, and JAX.
t5-static-cache
π€ Transformers: State-of-the-art Machine Learning for Pytorch, TensorFlow, and JAX.