[TRTLLM-14806][feat] Prefer Python V2 transceiver for LlamaForCausalLM and Gemma3ForCausalLM - #16787

[TRTLLM-14806][feat] Prefer Python V2 transceiver for LlamaForCausalLM and Gemma3ForCausalLM#16787

moraxu merged 10 commits into

NVIDIA:mainNVIDIA/TensorRT-LLM:mainfrom

moraxu:dev-mguzek-use-transceiver-v2-for-llama-and-gemmamoraxu/TensorRT-LLM:dev-mguzek-use-transceiver-v2-for-llama-and-gemmaCopy head branch name to clipboard