We’ve recently been working with mlc-llm and believe that it’s an excellent library.
However, the version of TVM currently used in MLC-LLM has diverged quite a bit from the mainline TVM. Additionally, there doesn’t seem to have been much development activity recently.
Considering that LLMs are one of the most important components in modern deep learning, we’re curious about the future plans for MLC-LLM.
Please note that this isn’t a critical question — we’re simply and genuinely interested in the direction of the project.