LFM2.5-2.6B
Release in the LFM (Liquid Foundation Models) family · version LFM2.5-2.6B
A 2.69B-parameter text-only hybrid model with 30 layers (22 double-gated short-convolution blocks and 8 grouped-query attention blocks), a 131,072-token context window, and support for 16 languages. Liquid AI pretrained it on about 34 trillion tokens and post-trained it for agentic use; it always reasons before answering.14
- Model hub: Model card (Hugging Face) (external site: huggingface.co)
- License: License (LFM Open License v1.0) (external site: huggingface.co)
- Release notes: Release post (external site: liquid.ai)
Availability and license
Overall availability
Weights download from Hugging Face without an access gate; use is governed by the LFM Open License v1.0, which conditions commercial-use rights on the licensee's legal entity not exceeding a threshold defined as annual revenue of US$10,000,000 or more.12
Availability is separate from permission: read the license before using or redistributing.
Custom Liquid AI license (card license field "lfm1.0") modeled on Apache 2.0, with the same text as the other LFM2.5 releases reviewed. Commercial rights are conditional on the licensee's legal entity not exceeding a "Threshold" defined as annual revenue of US$10,000,000 or more, and commercial use by a legal entity that exceeds the Threshold is not licensed; the Threshold does not apply to a qualified non-profit organization's use for non-commercial or research purposes. The license also requires redistributors to include it and mark modified files, and ends patent rights for anyone who files patent litigation over the work.23
Model-disclosure tier
The model parameters for this release can be downloaded by the public. License terms may still restrict use, redistribution, or commercial use.
The weights are under a license that is not on the rubric's OSI-approved list. Read its terms before use.
Public materials checklist
| Item | Status | Notes and evidence |
|---|---|---|
| WeightsCan the general public download the model parameters for this release? | Public | Ungated safetensors weights on Hugging Face; GGUF, ONNX, and MLX conversions and a base checkpoint are published as separate repositories.1 |
| Inference codeIs code for running the model published? | Public | The model card documents inference with Hugging Face Transformers (5.0.0 or later), vLLM, SGLang, llama.cpp, MLX, and LM Studio, and gives setup examples for several agent harnesses.1 |
| Training codeIs the code used to train the model published? | Unknown | No training code release was found in the model card or release post. |
| Training-data informationPublic = the training data itself can be obtained. Partial = composition or sources are documented without full access. | Unknown | The model card and release post state the pretraining budget (about 34 trillion tokens) but do not describe data sources or composition. |
| Training recipeAre the training configuration and procedure documented in enough detail to follow? | Partial | The model card and release post describe the stages at a high level: pretraining, a mid-training phase that extends context to 128K, two rounds of supervised fine-tuning, per-domain teacher specialization, multi-domain on-policy distillation, and agentic reinforcement learning inside agent harnesses. Hyperparameters and data mixes are not given.14 |
| Evaluation materialsPublic = evaluation code or prompts that let others re-run the evaluations are published. Partial = results only. | Partial | The model card reports a benchmark table against other small models; evaluation code or prompts are not linked.1 |
What it is useful for
The model card recommends it for agentic workloads, tool use, data extraction, retrieval-augmented generation, and long-context workflows, and does not recommend it for agentic coding or knowledge-heavy tasks.1
Run and use notes
- The model card recommends temperature 0.1, top_k 50, and repetition_penalty 1.1, and shows Transformers loading in bfloat16. It points to a separate 328M speculative-decoding drafter, LFM2.5-2.6B-DSpark, for SGLang and Apple silicon.1
Organization context
Provenance and derivatives
Post-trained by Liquid AI from its own LFM2.5-2.6B-Base checkpoint; no third-party base model is involved.1
- Derived from: LFM2.5-2.6B-Base (external site: huggingface.co) — Same organization; pretrained base checkpoint.
Other releases in the LFM (Liquid Foundation Models) family
- LFM2.5-8B-A1BModel-disclosure tier (USASI rubric v0.1): Open-weight
U.S. eligibility
Sources
This listing is not an endorsement, a safety assessment, or a federal approval.