You can join our ongoing works if you are interested

  • Apu 32B : An LLM for 23 south-asian langauges

    LLM Post-Training w/ syhthetic data
    Model Merging approaches
    Post Training Strategies
    Data curation

    Join >>

  • Do all tasks benefit equally from using reasoning and agents ?

    Reasoning approach VS task performace
    Task performance with/without agents
    Routing for models/reasoning/frameworks

    Join >>