alt.freeweights.lfm / LFM2.5-350M-ONNX
From: onnx-community <huggingface.co/onnx-community>Newsgroups: alt.freeweights.lfmSubject:

LFM2.5-350M-ONNX

Date: 2026-11-15Commit: 7dd4999Signed: 2DC28CC7B26A283B ✓

Text generation model. Onnx weights, 3.24 GB in all. Published by onnx-community under LFM-1.0 (custom).

lfm3.24GBLFM2.5-350M-ONNXonnxcustom terms
FactsParamsnot statedSize3.24 GBFormatsonnxTasktext generationLicenseLFM-1.0 (custom)From sourcecustom termsread terms
# custom terms: read the license before you download
$ fw get onnx-community/LFM2.5-350M-ONNX
# manifest sha256 bbbcac44bfd7… · 3.24 GB · 17 files
$ fw verify ./LFM2.5-350M-ONNX onnx-community/LFM2.5-350M-ONNX
From sourceHosted (signed)onnx-community/LFM2.5-350M-ONNXLast check: 2026-11-15 00:22 UTC

Get help: custom fine-tune or private deploy. We review your setup before we quote.

Quantized of LFM2.5-350M.

Works with

Runs with

Ranked by fit to this model (onnx), then by GitHub stars.

Convert with

Overview

Size
3.24 GB
Weights
255 MB (smallest of 5 ONNX variants)
Parameters
not stated
Formats
onnx
Task
text generation
License
LFM-1.0 (custom) (custom)
Files
17, each with a sha256
Publisher
onnx-community
Added
2026-11-15
Section
alt.freeweights.lfm
Lineage

Original release from the publisher.

Base model: LiquidAI/LFM2.5-350M

Where these bytes come from
Publisher at the pinned commit→You

Sources are tried in order. A file counts only when its size and sha256 match the manifest. Sources are checked when you ask for them, not on a timer.

Sources
BadgeHostLast checkResult
From sourcehuggingface.co2026-11-15 00:22 UTCpassed

Model card

From the publisher at commit 7dd4999. Shown as plain text with links; images open at the publisher.

Image: Liquid AI

Try LFM • Docs • LEAP • Discord

LFM2.5-350M#

LFM2.5 is a new family of hybrid models designed for on-device deployment. It builds on the LFM2 architecture with extended pre-training and reinforcement learning.

  • Best-in-class performance: A 350M model rivaling much larger models, bringing high-quality AI to your pocket.
  • Fast edge inference: 313 tok/s decode on AMD CPU, 188 tok/s on Snapdragon Gen4. Runs under 1GB of memory with day-one support for llama.cpp, MLX, and vLLM.
  • Scaled training: Extended pre-training from 10T to 28T tokens and large-scale multi-stage reinforcement learning.

Find more information about LFM2.5-350M in our blog post.

Image: image

🗒️ Model Details#

ModelParametersDescription
LFM2.5-350M-Base350MPre-trained base model for fine-tuning
LFM2.5-350M350MGeneral-purpose instruction-tuned model

LFM2.5-350M is a general-purpose text-only model with the following features:

  • Number of parameters: 350M
  • Number of layers: 16 (10 double-gated LIV convolution blocks + 6 GQA blocks)
  • Training budget: 28T tokens
  • Context length: 32,768 tokens
  • Vocabulary size: 65,536
  • Knowledge cutoff: Mid-2024
  • Languages: English, Arabic, Chinese, French, German, Japanese, Korean, Portuguese, Spanish
  • Generation parameters:
    • temperature: 0.1
    • top_k: 50
    • repetition_penalty: 1.05
ModelDescription
LFM2.5-350MOriginal model checkpoint in native format. Best for fine-tuning or inference with Transformers and vLLM.
LFM2.5-350M-GGUFQuantized format for llama.cpp and compatible tools. Optimized for CPU inference and local deployment with reduced memory usage.
LFM2.5-350M-ONNXONNX Runtime format for cross-platform deployment. Enables hardware-accelerated inference across diverse environments (cloud, edge, mobile).
LFM2.5-350M-MLXMLX format for Apple Silicon. Optimized for fast inference on Mac devices using the MLX framework.

We recommend using it for data extraction, structured outputs, and tool use. It is not recommended for knowledge-intensive tasks and programming.

Chat Template#

LFM2.5 uses a ChatML-like format. See the Chat Template documentation for details. Example:

<|startoftext|><|im_start|>system
You are a helpful assistant trained by Liquid AI.<|im_end|>
<|im_start|>user
What is C. elegans?<|im_end|>
<|im_start|>assistant

You can use tokenizer.apply_chat_template() to format your messages automatically.

Tool Use#

LFM2.5 supports function calling as follows:

  1. Function definition: We recommend providing the list of tools as a JSON object in the system prompt. You can also use the tokenizer.apply_chat_template() function with tools.
  2. Function call: By default, LFM2.5 writes Pythonic function calls (a Python list between <|tool_call_start|> and <|tool_call_end|> special tokens), as the assistant answer. You can override this behavior by asking the model to output JSON function calls in the system prompt.
  3. Function execution: The function call is executed, and the result is returned as a "tool" role.
  4. Final answer: LFM2 interprets the outcome of the function call to address the original user prompt in plain text.

See the Tool Use documentation for the full guide. Example:

<|startoftext|><|im_start|>system
List of tools: [{"name": "get_candidate_status", "description": "Retrieves the current status of a candidate in the recruitment process", "parameters": {"type": "object", "properties": {"candidate_id": {"type": "string", "description": "Unique identifier for the candidate"}}, "required": ["candidate_id"]}}]<|im_end|>
<|im_start|>user
What is the current status of candidate ID 12345?<|im_end|>
<|im_start|>assistant
<|tool_call_start|>[get_candidate_status(candidate_id="12345")]<|tool_call_end|>Checking the current status of candidate ID 12345.<|im_end|>
<|im_start|>tool
[{"candidate_id": "12345", "status": "Interview Scheduled", "position": "Clinical Research Associate", "date": "2023-11-20"}]<|im_end|>
<|im_start|>assistant
The candidate with ID 12345 is currently in the "Interview Scheduled" stage for the position of Clinical Research Associate, with an interview date set for 2023-11-20.<|im_end|>

🏃 Inference#

LFM2.5 is supported by many inference frameworks. See the Inference documentation for the full list.

NameDescriptionDocsNotebook
TransformersSimple inference with direct access to model internals.LinkColab link
vLLMHigh-throughput production deployments with GPU.LinkColab link
llama.cppCross-platform inference with CPU offloading.LinkColab link
MLXApple's machine learning framework optimized for Apple Silicon.Link—
LM StudioDesktop application for running LLMs locally.Link—

Here's a quick start example with Transformers.js:

If you haven't already, you can install the Transformers.js JavaScript library from NPM using:

npm i @huggingface/transformers

You can then use the model as follows:

import { pipeline, TextStreamer } from "@huggingface/transformers";

// Create a text generation pipeline
const generator = await pipeline(
  "text-generation",
  "onnx-community/LFM2.5-350M-ONNX",
  { dtype: "q4", device: "webgpu" },
);

// Define the list of messages
const messages = [
  { role: "system", content: "You are a helpful assistant." },
  { role: "user", content: "Tell me a story about a brave knight." },
];

// Generate a response
const output = await generator(messages, {
  max_new_tokens: 512,
  do_sample: false,
  streamer: new TextStreamer(generator.tokenizer, {
    skip_prompt: true,
    skip_special_tokens: true,
  }),
});
console.log(output[0].generated_text.at(-1).content);

See example output

Sure! Here's a story about a brave knight:

Once upon a time, in a quiet village, there lived a brave knight named Sir Cedric. Sir Cedric was known for his courage and unwavering bravery in battle. One day, during a fierce storm, the villagers were in dire need of help. The winds howled and the rain lashed down, but Sir Cedric led his men to safety, guiding them through the turbulent skies.

As the storm raged on, the villagers were desperate for food and shelter. Sir Cedric knew he had to act quickly to save their lives. He led his men to a hidden cave deep within the woods, where they could find warmth and safety.

When the storm finally passed, the villagers emerged from the cave, their hearts filled with relief and gratitude. Sir Cedric had saved their lives, and the villagers were grateful for his bravery. From that day on, Sir Cedric became a legendary figure in the village, remembered for his courage and selflessness.

Would you like to know more about Sir Cedric's story or perhaps explore another tale?

🔧 Fine-Tuning#

We recommend fine-tuning LFM2.5 for your specific use case to achieve the best results.

NameDescriptionDocsNotebook
CPT (Unsloth)Continued Pre-Training using Unsloth for text completion.LinkColab link
CPT (Unsloth)Continued Pre-Training using Unsloth for translation.LinkColab link
SFT (Unsloth)Supervised Fine-Tuning with LoRA using Unsloth.LinkColab link
SFT (TRL)Supervised Fine-Tuning with LoRA using TRL.LinkColab link
DPO (TRL)Direct Preference Optimization with LoRA using TRL.LinkColab link
GRPO (Unsloth)GRPO with LoRA using Unsloth.LinkColab link
GRPO (TRL)GRPO with LoRA using TRL.LinkColab link

📊 Performance#

Benchmarks#
ModelGPQA DiamondMMLU-ProIFEvalIFBenchMulti-IF
LFM2.5-350M30.6420.0176.9640.6944.92
LFM2-350M27.5819.2964.9618.2032.92
Granite 4.0-H-350M22.3213.1461.2717.2228.70
Granite 4.0-350M25.9112.8453.4815.9824.21
Qwen3.5-0.8B (Instruct)27.4137.4259.9422.8741.68
Qwen3.5-0.8B (Thinking)19.29-*32.9322.0026.44
Gemma 3 1B IT23.8914.0463.4920.3344.25
ModelCaseReportBenchBFCLv3BFCLv4τ²-Bench Telecomτ²-Bench Retail
LFM2.5-350M32.4544.1121.8618.8617.84
LFM2-350M11.6722.9512.2910.825.56
Granite 4.0-H-350M12.4443.0713.2813.746.14
Granite 4.0-350M0.8439.5813.732.926.14
Qwen3.5-0.8B (Instruct)13.8335.0818.7012.576.14
Qwen3.5-0.8B (Thinking)0.3939.6425.3914.337.02
Gemma 3 1B IT2.2816.617.179.366.43

*Evaluation could not be completed due to doom looping.

CPU Inference#

Image: image

GPU Inference#

Image: image

📬 Contact#

Citation#

@article{liquidAI2026350M,
  author = {Liquid AI},
  title = {LFM2.5-350M: No Size Left Behind},
  journal = {Liquid AI Blog},
  year = {2026},
  note = {www.liquid.ai/blog/lfm2-5-350m-no-size-left-behind},
}
@article{liquidai2025lfm2,
  title={LFM2 Technical Report},
  author={Liquid AI},
  journal={arXiv preprint arXiv:2511.23404},
  year={2025}
}

Files

17 files · 3.24 GBWeights 255 MB (smallest of 5 ONNX variants). Sizes and sha256 from the signed manifest.
Hide all 17 files with their sha256
FileSizesha256
.gitattributes1.83 KBba44193cf00eb237e2824cbfa212775d626f48f7c4caa3e630deda0908552db2
README.md15.2 KBd9b75b41642fd0c14ae5ed1f53a9978a474ecdd9078d6e45792cbe92de9f1d9a
chat_template.jinja2.55 KB013eed60546434b6967e3483153d8c5c37abcb1d667f8b1f914683f2a9411531
config.json1.44 KB544d8d604bacf4cb89383c49c9a54621afa26a6741f3f55fd8b840ca1d640419
generation_config.json136 B94bfac0e1c207691baf4e172389a8efb114f8b60eb3a5c07a2f418aefa8f8bb6
onnx/model.onnx141 KBccbf08c2e3a752af056182733da1ce9f0b0ad4961d1bc0fc97ef4a6f064c6e5b
onnx/model.onnx_data1.45 GBa6401d6753c3e299b0dbe7778489339c0aa3af480b2446d47c4d4a46b18c32d8
onnx/model_fp16.onnx140 KBfeb1c4a52e2756183f5ec7dd8d2c2d53d31465ba5050765b6c9f9c7a0a6c4705
onnx/model_fp16.onnx_data725 MB02f0ed790a23eafc73b1c5a77cf85628b4c04294fdc391aea387b58ad3b52f36
onnx/model_q4.onnx183 KBd1a705712e93aafaba1346b32245fa59e7857a46e2272003c0e8c524977e0de8
onnx/model_q4.onnx_data294 MB71ec6ad38a4c463dcb3dba671d06a1d9861be3a23e51290d818b95c0b7d2a5db
onnx/model_q4f16.onnx183 KB3012c28a119828561c90196331435d91c43c1a6ab4898ac79715585fff1dff85
onnx/model_q4f16.onnx_data255 MB9256ecd417b801b441d926ebe5ead4ebf50ac1a5a0f8f7914cf60d9d6452ef69
onnx/model_quantized.onnx186 KBbb7d0a2b14c7ffc8031d306e7d4dacb88f14070082c5708b673afe05013adde3
onnx/model_quantized.onnx_data510 MB6a37d3df8609169f802eb7ca3b8d19669bf9367f292ab46696c84bbe3b19cbd9
tokenizer.json3.30 MB29d43b4be8e8a896fefd7cd836ca6d6b4eedd249f823866ce0453b368e646f49
tokenizer_config.json3.83 KBe70cd070c2696bf1b0f1ad303c34502753670cca6f844637821cb1c37e6bea8d
VersionsEach time the pinned upstream commit or file list changed in this repo. Newest first.
Upstream commitDateContents
7dd4999565 pinned now2026-11-15 00:22 UTC17 files · 3.24 GB

Use it

# custom terms: read the license before you download
$ fw get onnx-community/LFM2.5-350M-ONNX
# manifest sha256 bbbcac44bfd7… · 3.24 GB · 17 files
$ fw verify ./LFM2.5-350M-ONNX onnx-community/LFM2.5-350M-ONNX
Terms that come with these files. LFM 1.0: commercial use only for organisations under US$10M annual revenue. Keep a copy of the license with the files.
Custom license (custom). Read the license text first. Open this panel to show the commands once you accept its terms.
fw.pyChecks the signature and every sha256, tries each source in order, resumes.
$ python3 fw.py get https://freeweights.allternit.com/manifests/onnx-community/LFM2.5-350M-ONNX.json --dir ./LFM2.5-350M-ONNX $ python3 fw.py verify ./LFM2.5-350M-ONNX ./onnx-community__LFM2.5-350M-ONNX.json
hfThe Hugging Face CLI with one setting. Same repo id.
$ HF_ENDPOINT=https://freeweights.allternit.com hf download onnx-community/LFM2.5-350M-ONNX --revision 7dd4999565b0342c381ba90a3d8fc467d6df19c4 --local-dir ./LFM2.5-350M-ONNX

License

Name
LFM Open License v1.0
Id
LicenseRef-LFM-1.0
Class
Custom terms
Redistribution
click-through
Listing
Hosted (signed)
Terms
LFM 1.0: commercial use only for organisations under US$10M annual revenue. Keep a copy of the license with the files.
Decided
License as declared by the publisher. No person read it. 2026-11-15
Text in
README.md
Text sha256
d9b75b41642fd0c14ae5ed1f53a9978a474ecdd9078d6e45792cbe92de9f1d9a
Terms gate
Yes. The commands stay folded until you open them after reading the license.

Read the license text

Licenses are checked against the license table. A custom license passes only after a person reads it.

Provenance

From: onnx-community <huggingface.co/onnx-community>Newsgroups: alt.freeweights.lfmSubject: LFM2.5-350M-ONNXDate: 2026-11-15Commit: 7dd4999Signed: 2DC28CC7B26A283B ✓
Upstream
onnx-community/LFM2.5-350M-ONNX
Pinned commit
7dd4999565b0342c381ba90a3d8fc467d6df19c4
Manifest
LFM2.5-350M-ONNX.json · .minisig
Manifest sha256
bbbcac44bfd74e34923ad59e270be742b4063001390113fa1a58a0f5a796f609
Signature
minisign, Ed25519, key 2DC28CC7B26A283B
Created
2026-11-15 00:22 UTC
Generator
freeweights-manifest/0.1.0
Takedowns
None.

How to verify

$ curl -O https://freeweights.allternit.com/manifests/onnx-community/LFM2.5-350M-ONNX.json $ curl -O https://freeweights.allternit.com/manifests/onnx-community/LFM2.5-350M-ONNX.json.minisig $ minisign -Vm LFM2.5-350M-ONNX.json -P RWQ7KGqyx4zCLXtUFOqvfv8L3g+8fpKJ1FsuxPyOYMF+4KufbUxKq1Oq $ shasum -a 256 <file> # compare with the Files tab

A manifest whose signature fails must be rejected. A file whose size or sha256 does not match must be discarded. Verify a download

Model tree

Parent

LiquidAI/LFM2.5-350M · this model is a quantization of it

No model in the catalog names this one as its parent.

Get it
Read license, then download
$ HF_ENDPOINT=https://freeweights.allternit.com hf download onnx-community/LFM2.5-350M-ONNX
Open in Freeweights Studio

Pull, verify and run it in the Allternit app on Desktop, web or phone.

Signed manifest
minisign, key 2DC28CC7B26A283B. LFM2.5-350M-ONNX.json. Provenance. Takedown history: none.
From: LiquidAI <huggingface.co/LiquidAI>Subject: LFM2-1.2B
From: LiquidAI <huggingface.co/LiquidAI>Subject: LFM2-1.2B-Extract
From: LiquidAI <huggingface.co/LiquidAI>Subject: LFM2-1.2B-RAG
From: LiquidAI <huggingface.co/LiquidAI>Subject: LFM2-1.2B-Tool
From: LiquidAI <huggingface.co/LiquidAI>Subject: LFM2-2.6B
From: LiquidAI <huggingface.co/LiquidAI>Subject: LFM2-2.6B-Exp