Revision History
NVIDIA TensorRT-LLM · 8 revisions
Sizes are character counts of the article source. The signed number is the change from the previous revision.
Recent edit summaries
Detailed summaries recorded by editors. Generic maintenance summaries are omitted here; every recorded revision remains below. Dates describe the edit, not necessarily the event it covers.
- Correction: Updated the article for the 2026 removal of TensorRT-LLM's TensorRT engine backend and Hugging Face's move of TGI to maintenance mode, and corrected wrong version attributions (Llama 3/Qwen2, Mistral 7B/Mixtral, PyTorch 25.10/ModelOpt, wide EP), an unsupported ONNX explanation of the PyTorch rearchitecture, an inverted speculative-decoding result, an unsupported 6x multi-LoRA figure, a contradicted vLLM latency claim, a recommendation misattributed to NVIDIA, and an incorrect claim that llama.cpp lacks continuous batching (plus 23 minor fixes)
Version 8 · Nov 2, 2026, 01:15 PM
- Update: added an official video.
Version 7 · Nov 1, 2026, 08:05 AM
- Improve internal links (2026-10-13)
Version 5 · Oct 13, 2026, 10:06 AM
- Added 6 contextual internal links
Version 4 · Jul 23, 2026, 03:46 PM