{"schema":"postcutoff/model@1","as_of":"2026-11-13T13:49:00+01:00","url":"https://postcutoff.com/m/liquid-lfm3-1-7b/","md":"https://postcutoff.com/m/liquid-lfm3-1-7b/index.md","disclosure":{"written_by":"AI agents (Claude Opus 5.5 in Claude Code)","editor":"human editor","policy":"https://postcutoff.com/about/"},"license":null,"id":"liquid-lfm3-1-7b","name":"LFM3-1.7B","org":"Liquid AI","family":"LFM3","released":"2026-10-27","status":"current","type":"llm","modality_in":["text"],"modality_out":["text"],"open_weights":true,"model_license":"lfm1.0","context_window":256000,"max_output":null,"knowledge_cutoff":null,"pricing":null,"price_line":"Price not published","access":[{"provider":"Hugging Face (text model; also LFM3-1.7B-Base, GGUF and MLX builds)","model_id":"LiquidAI/LFM3-1.7B","url":"https://huggingface.co/LiquidAI/LFM3-1.7B"},{"provider":"Hugging Face (siblings on the same backbone)","url":"https://huggingface.co/LiquidAI"}],"capabilities":[{"name":"Small on-device agent model","detail":"1.7B tri-hybrid model (13 gated short convolutions, 5 sliding-window and 6 global attention layers), ~1 GB at 4-bit, 256K context; Liquid reports WebArena-Verified 38.42 vs 3.02 for LFM2.5-2.6B and says it beats Qwen3.5-9B on long browser and business-app tasks (vendor numbers).","first":false,"discovered":"launch","source":"https://huggingface.co/blog/LiquidAI/lfm3"},{"name":"One backbone for every modality","detail":"Six 2B-class open models branch from LFM3-1.7B-Base: text, LFM3-VL-2B (first LFM with video input), LFM3-Audio-2B (18 languages, speech in and out), LFM3-Omni-2B (Liquid's first omni model) and the d1-2B / d1-Omni-2B decision models.","first":false,"discovered":"launch","source":"https://huggingface.co/blog/LiquidAI/lfm3"}],"entry":"2026-10-27-liquid-ai-lfm3-family","notes":"Released Oct 27, 2026 under the LFM Open License v1.0 (Hugging Face 'lfm1.0'). Pre-trained on 22T tokens with 3.5T tokens of mid-training on agentic traces; post-trained for agent harnesses such as Hermes Agent, Pi and OpenClaw; fine-tunes on one GPU. Benchmarks are Liquid's own.","verified":"2026-10-29","body_md":"Run with a recent `transformers` (the vision, audio and omni siblings need 5.20–5.21 and `trust_remote_code=True`); see the model\ncards for llama.cpp and MLX builds.","page_url":"https://postcutoff.com/m/liquid-lfm3-1-7b/","events_after":null,"major_after":null,"historic_after":null,"missing_at_launch":null,"events_since_release":365,"you_url":"https://postcutoff.com/you/liquid-lfm3-1-7b/","briefings":null}