Instructions to use SC117/LFM2.5-2.6B-Uncensored with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use SC117/LFM2.5-2.6B-Uncensored with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="SC117/LFM2.5-2.6B-Uncensored")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("SC117/LFM2.5-2.6B-Uncensored")
model = AutoModelForCausalLM.from_pretrained("SC117/LFM2.5-2.6B-Uncensored", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use SC117/LFM2.5-2.6B-Uncensored with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "SC117/LFM2.5-2.6B-Uncensored"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "SC117/LFM2.5-2.6B-Uncensored",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/SC117/LFM2.5-2.6B-Uncensored

SGLang

How to use SC117/LFM2.5-2.6B-Uncensored with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "SC117/LFM2.5-2.6B-Uncensored" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "SC117/LFM2.5-2.6B-Uncensored",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "SC117/LFM2.5-2.6B-Uncensored" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "SC117/LFM2.5-2.6B-Uncensored",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use SC117/LFM2.5-2.6B-Uncensored with Docker Model Runner:
```
docker model run hf.co/SC117/LFM2.5-2.6B-Uncensored
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

ABLITERIX TRIAL 65 BF16 LFM Open 1.0

LFM2.5-2.6B-Uncensored

English | 📖 中文文档

Uncensored 2.6B edge model · abliterix Trial 65 · BF16 safetensors

🌊 About this release

LFM2.5-2.6B is a Liquid AI 2.6B-parameter hybrid edge model built for agentic workloads: 30 layers (22 double-gated short-convolution blocks + 8 GQA), a 128K context window, 128K vocabulary, and a ChatML-like template with native <think> reasoning. It is competitive with models 4x larger on tool use, instruction following, and multi-step agentic tasks.

This release builds on the official weights in two steps:

Uncensored behavior edit via abliterix on ROCm (gfx1151), selecting Trial 65 LoRA and stream-merging it back to BF16.
Full-precision BF16 export — no quantization loss; quantized GGUFs (Q4_K_M / Q6_K / Q8_0 / IQ4_XS / IQ3_XS, imatrix-calibrated) ship in the GGUF sibling repo.

License: LFM Open License v1.0 (same as the base model). See LICENSE.

⚠️ Uncensored notice

After merging abliterix Trial 65, this model shows a much lower refusal rate and can differ substantially from official LFM2.5-2.6B. Evaluate compliance and safety for your use case; control access and audit as needed.

Refusals (harmful eval)	6 / 100 (baseline ~90 / 100)
KL divergence	0.0335 (same-prefix, far below 0.5 prune threshold)
Length deviation	0.079 σ
Generation health	PASSED
Selected trial	abliterix Trial 65
Thinking	Preserved — always-thinks (`<think>` in chat template)

Implementation sketch: LoRA merge W += (B @ A) * (alpha / r) (this trial alpha = r = 1); steering applied to attn.o_proj / conv.out_proj / mlp.down_proj across 30 layers.

🧠 Model details

Architecture	LFM2 hybrid (transformers `lfm2`, `Lfm2ForCausalLM`)
Parameters	2.69B total
Layers	30 (22 double-gated short-conv + 8 GQA)
Context	131,072 tokens
Vocab	128,000
Hidden / FFN	2048 / 10752
Reasoning	Always-thinks (`<think>` in chat template)
Languages	EN, ZH, AR, FR, DE, HI, ID, IT, JA, KO, PL, PT, RU, ES, TH, VI
This repo	BF16 safetensors (single shard, ~5.4 GB) + tokenizer + chat template

Coding / tool / agentic ability is largely retained; refusal and alignment behavior are changed. Official benchmark tables were not re-run for this derivative.

📦 What's in this repo

`model.safetensors`	Merged BF16 weights (5.39 GB)
`config.json` / `generation_config.json`	lfm2 config + official sampling defaults
`tokenizer.json` / `tokenizer_config.json`	128K vocab tokenizer
`chat_template.jinja`	ChatML-like template with `<think>` + tool-use tokens

🚀 Usage

The lfm2 architecture is natively supported by transformers >= 5.0.0 — no trust_remote_code needed.

from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "SC117/LFM2.5-2.6B-Uncensored"
model = AutoModelForCausalLM.from_pretrained(model_id, dtype="bfloat16")
tokenizer = AutoTokenizer.from_pretrained(model_id)
messages = [{"role": "user", "content": "What is 2+2?"}]
input_ids = tokenizer.apply_chat_template(
    messages, add_generation_prompt=True, return_tensors="pt"
)["input_ids"].to(model.device)
output = model.generate(
    input_ids,
    do_sample=True,
    temperature=0.1,
    top_k=50,
    repetition_penalty=1.1,
    max_new_tokens=512,
)
print(tokenizer.decode(output[0][input_ids.shape[1]:], skip_special_tokens=True))

llama.cpp / LM Studio users: use the quantized GGUFs in the GGUF sibling repo (Q4_K_M / Q6_K / Q8_0 / IQ4_XS / IQ3_XS, all imatrix-calibrated).

🎛️ Recommended sampling

Keep the official generation defaults: temperature 0.1, top_k 50, repetition_penalty 1.1 (they are baked into generation_config.json). If you want more creative answers, raise temperature toward 0.6–0.8; note the model always thinks before answering, so allow enough max_new_tokens for the <think> block.

🔧 Build pipeline

abliterix trial search on ROCm (gfx1151): 60 trials + 20 warmup, seed 117; all trials pruned only by same-prefix kl_divergence < 0.5.
Selected Trial 65: refusals 6/100 (baseline 90/100), KL 0.0335, length deviation 0.079 σ, generation health PASSED.
LoRA stream-merged into base weights in BF16 (W += B@A, alpha = r = 1).
BF16 GGUF converted (llama.cpp lfm2), then quantized with imatrix calibration (401 chunks from the APEX calibration set) into Q4_K_M / Q6_K / Q8_0 / IQ4_XS / IQ3_XS — see the GGUF repo.

Community derivative (behavior edit + BF16 release). Not an official Liquid AI release. Use at your own risk; follow local law and the LFM Open License v1.0.