Inference⚡
SGLang
Released Jan 18, 2024v0.1.5
Open-source serving engine for LLMs with RadixAttention, continuous batching and first-class speculative decoding (MTP, EAGLE, draft models, ngram). Apache-2.0 license.
LLMServingGPUHigh-throughput
Tools, frameworks, and models for local AI
1 resource found for "Coding"
Open-source serving engine for LLMs with RadixAttention, continuous batching and first-class speculative decoding (MTP, EAGLE, draft models, ngram). Apache-2.0 license.