Add rope_theta to rope_scaling for transformers 5.4+ compatibility (#19) b4caa84 nvidia-oliver-holworthy commited on 21 days ago
Update vLLM usage docs: remove config_vllm.json overwrite, relax version pin, and clarify minimal required flags (#18) cefc239 ybabakhin nvidia-oliver-holworthy commited on Mar 3
Add support for transformers 4.44 through 5.0+ (#16) 5be83c7 nvidia-oliver-holworthy commited on Feb 4
Remove the setting of _attn_implementation from llama_bidirectional_model (#3) e14eab1 nvidia-oliver-holworthy commited on Feb 4
Add vllm config and information (#11) ac8c77b verified ybabakhin radekosmulski-nvidia commited on Nov 20, 2025