Atomic-Germ/Qwen3.8-Distilled-1.2B-NPU2

IF YOU USE COMMUNITY QWEN MODELS DO NOT UPGRADE TO FLM v1.0.2+

Qwen3.8-Distilled-1.2B-NPU2

FastFlowLM Q4NX conversion of [`FastFlowLM/LFM2.5-1.2B-Thinking-NPU2`](https://huggingface.co/FastFlowLM/LFM2.5-1.2B-Thinking-NPU2) for AMD XDNA NPU inference.

This repository contains a quantized Q4NX port of the model, compiled for the FastFlowLM (FLM) runtime. It is not a GGUF file.

Source repository

Metadata from the upstream Hugging Face repository:

Usage

Install and run

This repository works with flm-add, a small installer that copies the model into the FastFlowLM user directory and registers the tag. It never modifies the system FastFlowLM install.

pip install flm-add or uv tool install flm-add

uv tool install flm-add
flm-add Atomic-Germ/Qwen3.8-Distilled-1.2B-NPU2 --tag qwen3.8-distilled:1.2b --family lfm2.5-tk
FLM_CONFIG_PATH="$HOME/.config/flm/model_list.json" FLM_XCLBIN_PATH="$HOME/.config/flm" flm run qwen3.8-distilled:1.2b

Files

FLM Bench

Tested on an AMD Ryzen AI 340 Framework 13 laptop.

Source model card

<div align="center"> <img src="https://cdn-uploads.huggingface.co/production/uploads/61b8e2ba285851687028d395/2b08LKpev0DNEk6DlnWkY.png" alt="Liquid AI" style="width: 100%; max-width: 100%; height: auto; display: inline-block; margin-bottom: 0.5em; margin-top: 0.5em;" /> <div style="display: flex; justify-content: center; gap: 0.5em; margin-bottom: 1em;"> <a href="https://playground.liquid.ai/"><strong>Try LFM</strong></a> • <a href="https://docs.liquid.ai/lfm"><strong>Documentation</strong></a> • <a href="https://leap.liquid.ai/"><strong>LEAP</strong></a> </div> </div>

LFM2.5-1.2B-Thinking

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 1.2B model rivaling much larger models, bringing high-quality AI to your pocket.
• —Fast edge inference: 239 tok/s decode on AMD CPU, 82 tok/s on mobile NPU. 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 in our blog post.

🗒️ Model Details

LFM2.5-1.2B-Thinking is a general-purpose text-only model with the following features:

• —Number of parameters: 1.17B
• —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
• —Languages: English, Arabic, Chinese, French, German, Japanese, Korean, Spanish
• —Generation parameters:
• —temperature: 0.1
• —top_k: 50
• —top_p: 0.1
• —repetition_penalty: 1.05

We recommend using it for agentic tasks, data extraction, and RAG. 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. 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. 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. 3.Function execution: The function call is executed, and the result is returned as a "tool" role.
4. 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.

Here's a quick start example with Transformers:

from transformers import AutoModelForCausalLM, AutoTokenizer, TextStreamer

model_id = "LiquidAI/LFM2.5-1.2B-Thinking"
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    device_map="auto",
    dtype="bfloat16",
#   attn_implementation="flash_attention_2" <- uncomment on compatible GPU
)
tokenizer = AutoTokenizer.from_pretrained(model_id)
streamer = TextStreamer(tokenizer, skip_prompt=True, skip_special_tokens=True)

prompt = "What is C. elegans?"

input_ids = tokenizer.apply_chat_template(
    [{"role": "user", "content": prompt}],
    add_generation_prompt=True,
    return_tensors="pt",
    tokenize=True,
).to(model.device)

output = model.generate(
    input_ids,
    do_sample=True,
    temperature=0.1,
    top_k=50,
    top_p=0.1,
    repetition_penalty=1.05,
    max_new_tokens=512,
    streamer=streamer,
)

🔧 Fine-Tuning

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

📊 Performance

Benchmarks

We compared LFM2.5-1.2B-Thinking with relevant sub-2B models on a diverse suite of benchmarks.

GPQA, MMLU-Pro, IFBench, and AIME25 follow ArtificialAnalysis's methodology. For IFEval and Multi-IF, we report the average score across strict and loose prompt and instruction accuracies. For BFCLv3, we report the final weighted average score with a custom Liquid handler to support our tool use template.

Inference speed

LFM2.5-1.2B-Thinking offers extremely fast inference speed on CPUs with a low memory profile compared to similar-sized models.

In addition, we are partnering with AMD, Qualcomm, Nexa AI, and FastFlowLM to bring the LFM2.5 family to NPUs. These optimized models are available through our partners, enabling highly efficient on-device inference.

Prefill Performance

We report prefill throughput evaluated over a range of prompt lengths.

Decode Performance

The reported results correspond to decoding 100 tokens at different context lengths.

LFM2.5-1.2B-Thinking excels at long-context inference. On AMD NPUs with FastFlowLM, decoding throughput sustains ~46 tok/s even at the full 32K context, indicating robust long-context scalability. See detailed benchmark results (up to full context length) here.

These capabilities unlock new deployment scenarios across various devices, including vehicles, mobile devices, laptops, IoT devices, and embedded systems.

Contact

For enterprise solutions and edge deployment, contact sales@liquid.ai.

Citation

@article{liquidai2025lfm2,
  title={LFM2 Technical Report},
  author={Liquid AI},
  journal={arXiv preprint arXiv:2511.23404},
  year={2025}
}