Archive Show All6 By MTC Team5 New Feature3 Release Notes2 Research1 2025 Sep 04Accelerating Token Generation with MTP (Multi-Token Prediction) Sep 03LightLLM v1.1.0: Now Available! Jun 15Pre$^3$: Unlocking Faster, Structured LLM Generation with Deterministic Pushdown Automata Feb 16LightLLM v1.0.0: Now Available! Jan 22Reducing Overhead with Cuda Graph Jan 21Welcome To the LightLLM Blog