otherllmRuns locally1.73B (card)
LFM3-1.7B: Specs, License & How to Run
Liquid AI 1.73B edge text LLM, 256K. Not LFM3-Omni-2B and not LFM2.5.
- License
- other
- Parameters
- 1.73B (card)
- HF likes
- 635 (2026-11-01)
- First seen
- 2026-10-23
TL;DR. LFM3-1.7B is a llm model from Liquid AI. Parameters: 1.73B (card).License: other. It is listed under ChatGPT, Closed LLM APIs.
| Developer | Liquid AI |
|---|---|
| Type | llm |
| Architecture | lfm3 / Lfm3ForCausalLM. Card: 24 layers (13 gated short convolutions, 5 sliding-window attention with a 256-token window, 6 global attention); vocab 128,000. Text-only lead card in the LFM3 family. Distinct from LFM3-Omni-2B (2.17B any-to-any, 32K) and from LFM2.5-24B-A2B (24.0B / 2.45B text MoE). Card: not recommended for agentic coding or complex multi-step reasoning because of its small size. |
| Parameters | 1.73B (card) |
| Modalities | text |
| Languages | English, Arabic, Chinese, French, German, Italian, Japanese, Korean, Portuguese, Spanish, Vietnamese, Thai, Indonesian, Hindi, Russian, Polish |
| Context | 256K tokens |
| License | other — LICENSE on LiquidAI/LFM3-1.7B is LFM Open License v1.0 (cardData.license other / license_name lfm1.0). Licensor is Liquid AI, Inc. Section 5 conditions Commercial Use on You/Legal Entity not exceeding annual revenue Threshold of USD 10,000,000. Above the threshold, Commercial Use is not licensed under this agreement. Qualified Non-Profits using the work for non-commercial or research purposes are exempt from the threshold. Same grant on the GGUF / MLX / ONNX / FP8 siblings we opened. Not legal advice. |
| Hardware | Not yet verified — README names BF16, FP8, NVFP4, GGUF (llama.cpp / Ollama / LM Studio), MLX-4bit, and ONNX siblings. Same README says 4-bit precision “takes about 1 GB of memory.” Vendor, not a forkalt measurement. HF usedStorage on the BF16 card is 3,475,757,943 bytes (~3.24 GiB). (not independently tested) |
| Hugging Face likes | 635 (as of 2026-11-01) |
| Model card | https://huggingface.co/LiquidAI/LFM3-1.7B |
| First seen | 2026-10-23 |
LFM3-1.7B is Liquid AI’s text-only lead card in the LFM3 family: 1.73B parameters, 256K context, on-device. Official hosts we opened on 2026-11-01: the ungated card LiquidAI/LFM3-1.7B (created 2026-10-23T21:35:42Z; last modified 2026-10-28T17:16:37Z; 635 likes / 32,933 downloads), siblings LFM3-1.7B-GGUF (157 likes / 129,608 downloads), LFM3-1.7B-MLX-4bit (17 likes / 6,810 downloads), LFM3-1.7B-ONNX (13 likes / 1,078 downloads), and LFM3-1.7B-FP8 (12 likes / 1,076 downloads), the LFM Open License v1.0, the LFM3 collection, and the family post Introducing LFM3 (27 October 2026). API gated was false.
It is not LFM3-Omni-2B (2.17B any-to-any, 32K) and not LFM2.5-24B-A2B (24.0B / 2.45B text MoE). Head-to-head facts: LFM3-1.7B vs LFM3-Omni-2B and LFM3 vs LFM2.5-24B-A2B.
This identification page hangs off closed LLM APIs and ChatGPT alternatives. That mapping is editorial for a local text checkpoint. We have no sourced quality comparison against ChatGPT.
The text card says LFM3-1.7B is not recommended for agentic coding or complex multi-step reasoning because of its small size. That sentence stays vendor.
What we verified (2026-11-01)
| Field | Value | Source |
|---|---|---|
| Host | huggingface.co/LiquidAI/LFM3-1.7B (not gated) | HF API |
| Created / last modified | 2026-10-23T21:35:42Z / 2026-10-28T17:16:37Z | same |
| Likes / downloads | 635 / 32,933 | same |
| License tag + file | other / lfm1.0 + LFM Open License v1.0 |
HF cardData + LICENSE |
| Pipeline tag | text-generation | HF API |
| Architecture class | Lfm3ForCausalLM |
HF config |
| Params (card) | 1.73B | README “Model details” |
| safetensors.parameters | 1,728,801,536 | HF API |
| usedStorage (BF16) | 3,475,757,943 bytes (~3.24 GiB) | same |
| Context | 256K | README |
| GGUF sibling | 157 likes / 129,608 downloads | HF API |
Languages on the card: EN, AR, ZH, FR, DE, IT, JA, KO, PT, ES, VI, TH, ID, HI, RU, PL. Family overview (VL / Audio / Omni / d1) stays on LFM3.
How to run (from the official README, not tested)
Copied names only. forkalt has not loaded a checkpoint.
The README lists BF16 (vLLM, SGLang, Transformers), FP8, NVFP4, GGUF (llama.cpp, Ollama, LM Studio), MLX-4bit, and ONNX. Sampling copied from that README: general/chat temperature=0.1, top_k=50, repetition_penalty=1.1; math/STEM temperature=1.0, top_p=0.95, top_k=20. The card says 4-bit “takes about 1 GB of memory.” Vendor, not our measurement.
Hosted Liquid Playground is a different artifact from the weight grant.
Vendor benches (self-reported)
The LFM3 blog reports LiveClaw 45.60 and AutomationBench (public) 41.29 for LFM3-1.7B versus other open models. Those rows are self-reported. They are not a ChatGPT comparison and not a forkalt bake-off against LFM3-Omni-2B.
License
LICENSE is LFM Open License v1.0. Licensor is named Liquid AI, Inc. Section 5: Commercial Use rights are conditioned on You or Your Legal Entity not exceeding the Threshold of USD 10,000,000 annual revenue. Commercial Use by an entity over that threshold is not licensed under this agreement. Qualified Non-Profit Organizations using the work for Non-Commercial or Research Purposes are exempt from the threshold. Not legal advice.
Related on this site: LFM3 (family) — LFM3-Omni-2B — LFM3-1.7B vs LFM3-Omni-2B — LFM2.5-24B-A2B — LFM3 vs LFM2.5-24B-A2B — Spark-X2.6-4B — Granite Switch 4.1 — humanizer (style rewrite, different job). Serving: vLLM. Catalog: home. Corrections: contact.
Benchmarks
Quoted from the cited source. forkalt has not independently reproduced these numbers.
| Name | Score | Source |
|---|---|---|
| LiveClaw (vendor LFM3 blog) | 45.60 (self-reported) | https://www.liquid.ai/blog/lfm3 |
| AutomationBench public (vendor LFM3 blog) | 41.29 (self-reported) | https://www.liquid.ai/blog/lfm3 |
Deploy options
Comparisons
FAQ
Is LFM3-1.7B the same as LFM3-Omni-2B or LFM2.5-24B-A2B?
No. LFM3-1.7B is the text-only 1.73B / 256K lead card. LFM3-Omni-2B is the 2.17B any-to-any sibling (text/image/video/audio in; text+speech out; 32K). LFM2.5-24B-A2B is the earlier 24.0B / 2.45B text MoE. Same LFM Open License v1.0 on the LFM3 cards; different job and size. See LFM3-1.7B vs LFM3-Omni-2B and LFM3 vs LFM2.5-24B-A2B.
Are the weights public?
Yes. LiquidAI/LFM3-1.7B is ungated (created 2026-10-23T21:35:42Z; last modified 2026-10-28T17:16:37Z; 635 likes / 32,933 downloads in the 2026-11-01 API check). Siblings LFM3-1.7B-GGUF (157 likes / 129,608 downloads), LFM3-1.7B-MLX-4bit (17 likes), LFM3-1.7B-ONNX (13 likes), and LFM3-1.7B-FP8 (12 likes) were also ungated. Collection: huggingface.co/collections/LiquidAI/lfm3.
Can I use it commercially?
The LICENSE file is LFM Open License v1.0. Section 5 conditions Commercial Use on You/Legal Entity not exceeding USD 10,000,000 annual revenue. Above that threshold, Commercial Use is not licensed under this agreement. Qualified Non-Profits for non-commercial/research purposes are exempt from the threshold. Not legal advice.
Does it beat ChatGPT or LFM3-Omni-2B?
We do not claim that. Blog tables (LiveClaw 45.60, AutomationBench 41.29) are vendor self-reported. There is no sourced bake-off against named closed APIs or against the Omni sibling on this page.
Sources & verification
- https://huggingface.co/LiquidAI/LFM3-1.7B
- https://huggingface.co/LiquidAI/LFM3-1.7B/raw/main/LICENSE
- https://huggingface.co/LiquidAI/LFM3-1.7B-GGUF
- https://huggingface.co/LiquidAI/LFM3-1.7B-MLX-4bit
- https://huggingface.co/LiquidAI/LFM3-1.7B-ONNX
- https://huggingface.co/LiquidAI/LFM3-1.7B-FP8
- https://huggingface.co/LiquidAI/LFM3-Omni-2B
- https://huggingface.co/collections/LiquidAI/lfm3
- https://www.liquid.ai/blog/lfm3
License notes are not legal advice. Read the original LICENSE. Trademark & disclaimer.