Wispr Flow - Donโt type, just speak
Mini toolsLFM models
Browse all models from this model family.
ID
Model
Company
Type
Primary task
Open source
-
By Liquid AILFM2.5-230M is a 230M-parameter lightweight language model built on the LFM2 architecture, designed for fast inference across edge and cloud environments. Pre-trained on 19T tokens with a 32K context window, it targets tool use, data extraction, and agentic workflows. Runs on CPUs including Raspberry Pi 5 and mobile SoCs. Not recommended for math, code generation, or creative writing.NewTextReleased 1d ago
-
By Liquid AI350M-parameter bidirectional multilingual bi-encoder that maps each document to a single dense vector for fast, cost-efficient multilingual and cross-lingual search. Supports 11 languages. Built from LFM2.5-350M-Base. Produces the smallest, cheapest index among LFM2.5 retrieval models. Available in GGUF format for CPU and edge deployment.NewTextReleased 8d ago
-
By Liquid AI350M-parameter bidirectional multilingual retrieval model using per-token ColBERT-style late interaction (MaxSim) for accurate multilingual and cross-lingual search. Supports 11 languages. Built from LFM2.5-350M-Base with bidirectional attention and non-causal convolutions. Available in GGUF format for CPU and edge deployment.NewTextReleased 8d ago
-
By Liquid AILFM2.5-1.2B-Thinking is Liquid AIโs 1.2B parameter reasoning-focused variant of its LFM2.5 edge family, optimized for agentic workflows, RAG and data extraction with 32K context, and exposed on OpenRouter as a free model.TextReleased 5mo ago
-
By Liquid AILFM2.5-1.2B is Liquid AIโs 1.2B hybrid edge model for on-device chat and tools, trained on 28T tokens and RL, running under 1 GB of RAM while rivaling much larger LLMs on reasoning and agentic tasks.TextReleased 5mo ago
-
By Liquid AILFM2-VL-3B is a 3B vision-language model that reads images with text and answers in natural language or structured JSON. It handles OCR, charts, tables, and screenshots with long context and low-latency streaming, making it practical for multimodal RAG and assistants.TextReleased 8mo ago
-
By Liquid AILFM2-Audio-1.5 is a unified audio foundation model for real-time speech, music, and sound effects. It transcribes, speaks, and generates or edits audio from natural prompts, supports cross-lingual use, returns stable timestamps, and streams low-latency output for production apps.MultimodalReleased 8mo ago
