Model library / LiquidAI
LFM3-1.7B is a 1.7 billion-parameter text model for local agents and specialised applications.
Parameters in total
Context length in tokens, as published
Weights at the released precision
Licence declared on the repository
Overview
LFM3-1.7B is a text model released by LiquidAI, built on a tri-hybrid architecture and described as suited for local agents, tool calling, structured outputs and domain-specific fine-tuning. The publisher notes it is post-trained for agentic harnesses such as Hermes Agent, Pi and OpenClaw, and is not intended for heavy coding or complex multi-step reasoning due to its size.
In practice the model occupies roughly 1 GB of memory when run at 4-bit precision, fitting on typical phones or laptops. It supports a 262,144-token context window and is distributed under an "other" licence, with weights stored in bfloat16 format. Deployment can use vLLM, SGLang or compatible llama.cpp builds.
Specification
| Repository | LiquidAI/LFM3-1.7B |
| Publisher | LiquidAI |
| Published | 23 October 2026 |
| Architecture | Lfm3ForCausalLM |
| Model type | lfm3 |
| Parameters | 1,728,801,536 |
| Layers | 24 |
| Hidden size | 2,048 |
| Attention heads | 16, 4 key/value heads |
| Context length | 262,144 tokens |
| Vocabulary | 128,000 tokens |
| Weight format | bfloat16 |
| Licence | other |
| Base model | LiquidAI/LFM3-1.7B-Base |
Memory
Parameter count multiplied by the bytes each format uses. The key-value cache comes on top.
| Precision | Weights | Notes |
|---|---|---|
| bf16 | 3.2 GiB | Full precision as released |
| fp8 | 1.6 GiB | 8-bit, near-lossless on most models |
| int4 | 0.8 GiB | 4-bit, smallest footprint |
The cache costs 48 KiB per token at 16-bit, so the full 262,144-token context of one request needs about 12.0 GiB. Concurrency multiplies that number, not the weights.
That figure is an upper bound: this model uses sliding-window attention with a 256-token window on some layers, so long requests cache less than this.
Hardware
At the released precision, with 90% of the card given to the server and 2.5 GiB kept for the runtime. The context column is what is left for the cache on that setup.
| GPU | Cards needed | Free for cache | Context it holds |
|---|---|---|---|
| RTX 4090 24GB | 1 | 15.9 GiB | the full 262,144 tokens |
| RTX 5090 32GB | 1 | 23.1 GiB | the full 262,144 tokens |
| L40S 48GB | 1 | 37.5 GiB | the full 262,144 tokens |
| A100 80GB | 1 | 66.3 GiB | the full 262,144 tokens |
| H100 80GB | 1 | 66.3 GiB | the full 262,144 tokens |
| RTX PRO 6000 96GB | 1 | 80.7 GiB | the full 262,144 tokens |
| H200 141GB | 1 | 121.2 GiB | the full 262,144 tokens |
Serving
A vLLM launch line for the shape above. Check the model card for a runtime the publisher recommends.
vllm serve LiquidAI/LFM3-1.7B \ --max-model-len 262144 \ --gpu-memory-utilization 0.90
Questions
The weights take about 3 GiB at the released precision. That fits on one RTX 4090 24GB, which leaves roughly 16 GiB for the key-value cache.
The repository declares other. Read the licence text before commercial use: the name alone does not tell you what is allowed.
The configuration allows 262,144 tokens. Whether the whole window is usable depends on the memory left for the cache, which the hardware table works out per card.
This page is a reference, and Aiqre does not serve this model today. We deploy open-weight models on dedicated EU hardware on request, so if you want an OpenAI-compatible endpoint for it, email hello@aiqre.com with the model name and roughly what volume you expect.
Also in the library
LFM3-1.7B-Base is a 1.7 billion-parameter pre-trained language model from LiquidAI intended for heavy fine-tuning and specialised…
LFM2.5-24B-A2B is a 24 billion-parameter MoE reasoning model with 2.45 billion active parameters for on-device agentic tasks.
LFM2.5-24B-A2B-Base is a 24 billion-parameter mixture-of-experts model from LiquidAI aimed at on-device deployment and heavy fine…
LFM2.5-2.6B is a 2.7B open-weight model published by LiquidAI on Hugging Face. It uses the Lfm2ForCausalLM architecture with 30 l…
Hosting
Tell us the model and roughly what volume you expect. We reply with a price and an OpenAI-compatible endpoint, with your prompts never stored.