Liquid AI (Vendor): news timeline & CVEs

Liquid AI Releases d1: A Decision Model That Returns Calibrated Probabilities With Zero Output Tokens

Liquid AI launched d1, a hosted decision model returning calibrated probabilities with zero output tokens.

Liquid AI released d1, a hosted decision model that answers typed questions with calibrated probabilities and zero generated tokens. Its primitives are Noul for yes/no probabilities, Choice for a labeled option distribution, and Score for rubric positions, all evaluated in one call to the Liquid API as d1:free. Liquid says there are no downloadable weights and the model is not trainable. The company contrasts it with generative LLMs and with decision models such as TypeSafe Jev 1.13, Convai Laya, and JEV-27B.

MarkTechPostSep 29, 21:47 UTC · 1d agoModel release

Liquid AI Releases LFM2.5-VL-3B-DSpark: Speculative Decoding for Vision-Language Models With Up to 3.13x Faster Decoding

Liquid AI released a 280M drafter that speeds LFM2.5-VL-3B decoding by up to 3.13x.

Liquid AI released LFM2.5-VL-3B-DSpark, an experimental 279.5-million-parameter speculative-decoding drafter for its LFM2.5-VL-3B vision-language model. The four-layer drafter proposes tokens that the target verifies, adding about 8.9% parameters without changing greedy outputs. Liquid AI reports decode speedups up to 3.13x on an Apple M5 Max and 2.66x on an NVIDIA H100, with smaller end-to-end gains because image encoding and prefill are unchanged. Weights are on Hugging Face in Safetensors and GGUF, with support in SGLang, MLX-VLM, and llama.cpp, under a license that allows free commercial use only for companies under $10 million in annual revenue.

MarkTechPostSep 25, 23:11 UTC · 5d agoModel release

Liquid AI

Timeline2 articles

Appears with

Liquid AI Releases d1: A Decision Model That Returns Calibrated Probabilities With Zero Output Tokens

Liquid AI Releases LFM2.5-VL-3B-DSpark: Speculative Decoding for Vision-Language Models With Up to 3.13x Faster Decoding