Instructions to use saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic:Q4_K_M

Use Docker

docker model run hf.co/saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic:Q4_K_M

LM Studio
Jan

vLLM

How to use saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic:Q4_K_M

Ollama
How to use saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic with Ollama:
```
ollama run hf.co/saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic:Q4_K_M
```

Unsloth Studio

How to use saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic to start chatting

How to use saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic with Docker Model Runner:
```
docker model run hf.co/saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic:Q4_K_M
```

Lemonade

How to use saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic:Q4_K_M

Run and chat with the model

lemonade run user.lfm2.5-2.6b-fable5-coding-agent-heretic-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic:Q4_K_M

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic:Q4_K_M

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic:Q4_K_M" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

LFM2.5-2.6B-Fable5-Coding-Agent-heretic

A decensored variant of AyoubChLin/lfm2.5-2.6b-fable5-coding-agent (full-parameter SFT of LiquidAI/LFM2.5-2.6B on saidutta69/fable-5-premium), produced with Heretic v1.4.0 (directional ablation / "abliteration"). Refusal behavior is suppressed via targeted weight edits to the attention output and MLP down-projections rather than fine-tuning, so the base model's coding-agent capabilities, tool-use patterns, and instruction-following are left largely intact.

Abliteration results: KL divergence 0.014 · Refusals reduced from 96/100 → 7/100.

Who this is for: developers who want a compact 2.6B coding agent with LFM2's hybrid conv+attention architecture — fast inference, tool-call generation, code generation, and multi-turn assistant behavior — without refusal guardrails. Not a capability upgrade over the base model — same model, refusal guardrails removed.

Why abliteration instead of fine-tuning

Fine-tuning a "helpful" persona on top of RLHF'd refusals fights the base model's training and tends to degrade coherence. Abliteration instead finds and edits the specific weight directions responsible for refusal, leaving the rest of the network (and its capabilities) untouched. See the Heretic repo and the original abliteration writeup for the mechanism.

Files

GGUF quantizations

Full quantization set (14 quants + F16) produced with llama.cpp.

File	Format	Size
`lfm2.5-2.6b-fable5-coding-agent-heretic-F16.gguf`	GGUF F16	5.03 GB
`lfm2.5-2.6b-fable5-coding-agent-heretic-Q2_K.gguf`	GGUF Q2_K	1.02 GB
`lfm2.5-2.6b-fable5-coding-agent-heretic-IQ3_S.gguf`	GGUF IQ3_S	1.18 GB
`lfm2.5-2.6b-fable5-coding-agent-heretic-Q3_K_S.gguf`	GGUF Q3_K_S	1.18 GB
`lfm2.5-2.6b-fable5-coding-agent-heretic-Q3_K_M.gguf`	GGUF Q3_K_M	1.27 GB
`lfm2.5-2.6b-fable5-coding-agent-heretic-Q3_K_L.gguf`	GGUF Q3_K_L	1.35 GB
`lfm2.5-2.6b-fable5-coding-agent-heretic-IQ4_XS.gguf`	GGUF IQ4_XS	1.42 GB
`lfm2.5-2.6b-fable5-coding-agent-heretic-Q4_K_S.gguf`	GGUF Q4_K_S	1.49 GB
`lfm2.5-2.6b-fable5-coding-agent-heretic-Q4_0.gguf`	GGUF Q4_0	1.48 GB
`lfm2.5-2.6b-fable5-coding-agent-heretic-Q4_1.gguf`	GGUF Q4_1	1.63 GB
`lfm2.5-2.6b-fable5-coding-agent-heretic-Q4_K_M.gguf`	GGUF Q4_K_M	1.56 GB
`lfm2.5-2.6b-fable5-coding-agent-heretic-Q5_K_S.gguf`	GGUF Q5_K_S	1.77 GB
`lfm2.5-2.6b-fable5-coding-agent-heretic-Q5_K_M.gguf`	GGUF Q5_K_M	1.81 GB
`lfm2.5-2.6b-fable5-coding-agent-heretic-Q6_K.gguf`	GGUF Q6_K	2.07 GB
`lfm2.5-2.6b-fable5-coding-agent-heretic-Q8_0.gguf`	GGUF Q8_0	2.68 GB

LFM2 hybrid conv+attention architecture — loads natively in llama.cpp (arch lfm2).

Run llama serve -hf saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic to pull the default quant.

Quickstart

llama.cpp

# defaults to the Q4_K_M quant
llama serve -hf saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic:Q4_K_M

Ollama

ollama run hf.co/saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic:Q4_K_M

LM Studio

Open LM Studio and click the search icon to open the Model Search panel.
Type "lfm2.5-2.6b-fable5-coding-agent-heretic" and click the download button marked GGUF.
Pick your quant, load the model, and start chatting.

Transformers

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

MODEL_ID = "saidutta69/lfm2.5-2.6b-fable5-coding-agent-heretic"

tokenizer = AutoTokenizer.from_pretrained(MODEL_ID)
model = AutoModelForCausalLM.from_pretrained(
    MODEL_ID,
    dtype=torch.bfloat16,
    device_map="auto",
)

messages = [
    {"role": "system", "content": "You are a helpful coding assistant."},
    {"role": "user", "content": "Write a Python function that merges overlapping intervals."},
]

inputs = tokenizer.apply_chat_template(
    messages, tokenize=True, add_generation_prompt=True, return_tensors="pt"
).to(model.device)

with torch.inference_mode():
    output = model.generate(**inputs, max_new_tokens=512, temperature=0.1)

print(tokenizer.decode(output[0, inputs["input_ids"].shape[1]:], skip_special_tokens=True))

Responsible use

Refusal suppression is deliberate and works as intended: this model will comply with requests the base model would refuse, including some it shouldn't. There is no safety filtering layered on top. You are responsible for how you deploy it.