Liquid AI releases RIPL engine
The story so far
Liquid AI has released the millisecond-level real-time intelligent engine RIPL, which can return complete generation responses in less than 50 milliseconds and supports real-time requests on NVIDIA GPUs. It is claimed that Liquid RIPL reduces the p95 latency of LFM2.5-1.2B on RIPL by 14.4–28.4 times compared to Qwen3.5-0.8B on vLLM. Currently, Liquid RIPL is deployed in production at Shopify for search enhancement and is suitable for enterprise deployment on NVIDIA GPUs (such as H100, H200, B200, or RTX PRO 6000) for LFMs. Liquid AI states that RIPL is a low-latency inference engine designed to place Liquid Foundation Models (LFMs) directly in the real-time request path and can continuously process traffic under peak loads.
AI-written from the reports · Updated 27 days ago
Coverage timeline
Follow the coverage to see every side of the story.
- MarkTechPostLiquid AI Launches RIPL, a Real-Time Inference Engine for Sub-50 ms LFMs
Liquid AI has launched the RIPL real-time inference engine, serving the Liquid Foundation Model (LFM) in the request path, generating responses in less than 50 milliseconds end-to-end, and maintaining low latency under sustained peak traffic.
- Liquid AI BlogRIPL: Real-time Intelligence When Every Millisecond Counts
Liquid RIPL returns complete generated responses in less than 50 milliseconds, supporting real-time requests on NVIDIA GPUs. LFM2.5-1.2B shows a 14.4–28.4 times lower p95 latency on RIPL compared to Qwen3.5-0.8B on vLLM, and Shopify has deployed it in production for search enhancement.