Liquid AI launches sub-20ms latency stack

Liquid AI announced Liquid RIPL (Real-time Inference Pipeline), a low-latency inference engine that returns complete generated model responses end-to-end in under 50 ms while sustaining continuous peak-load traffic. Liquid AI reported that LFM2.5-1.2B on Liquid RIPL achieves 14.4–28.4× lower p95 latency than Qwen3.5-0.8B on vLLM across two real-time workloads, and LFM2.5-2.6B achieves 5.6–7.7× lower p95 latency than Qwen3.5-2B. The engine targets enterprise real-time workloads such as search ranking, product recommendations, fraud decisions and payment authorizations that have millisecond latency budgets.

News https://www.liquid.ai/news RIPL: Real-time Intelligence When Every Millisecond Counts Today we're announcing Liquid RIPL Real-time Inference Pipeline , a low-latency inference engine built to put Liquid Foundation Models LFMs directly in the path of a live request. It returns complete generated model responses end-to-end in under 50 ms , and it keeps doing so while traffic arrives continuously at peak load. Most of today's serving infrastructure is built for a different job: general purpose models, large batches, and long running agentic workflows where a few seconds per request is acceptable. Enterprise real-time workloads have far smaller latency budgets. Search rankings, product recommendations, fraud decisions, payment authorizations, and other real-time workloads need an answer in milliseconds. Even small delays add friction to the user experience, like a slower checkout that hurts conversion. That's why these systems have traditionally used fast, but less accurate, methods like gradient-boosted trees or small embedding models. Liquid’s RIPL changes that trade off. It’s built for real-time constraints, delivering faster, more responsive generative AI experiences that improve how customers search, discover, transact, and interact. Results across real-time workloads In Figure 1, we highlight a few different use cases for small LLMs in real time enterprise workflows, enabled by low-latency LLM inference. Across the two workloads in Figure 2, LFM2.5-1.2B on Liquid RIPL has 14.4–28.4× lower p95 latency than Qwen3.5-0.8B on vLLM. LFM2.5-2.6B has 5.6–7.7× lower p95 latency than Qwen3.5-2B.