LFM2.5-2.6B-UNCENSORED-ABLITERATED-PHILADELPHIA-CLASS

Instructions to use Justbackup/LFM2.5-2.6B-UNCENSORED-ABLITERATED-PHILADELPHIA-CLASS with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Justbackup/LFM2.5-2.6B-UNCENSORED-ABLITERATED-PHILADELPHIA-CLASS with Transformers:

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

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

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Justbackup/LFM2.5-2.6B-UNCENSORED-ABLITERATED-PHILADELPHIA-CLASS")
model = AutoModelForCausalLM.from_pretrained("Justbackup/LFM2.5-2.6B-UNCENSORED-ABLITERATED-PHILADELPHIA-CLASS", 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

llama.cpp

How to use Justbackup/LFM2.5-2.6B-UNCENSORED-ABLITERATED-PHILADELPHIA-CLASS 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 Justbackup/LFM2.5-2.6B-UNCENSORED-ABLITERATED-PHILADELPHIA-CLASS:BF16
# Run inference directly in the terminal:
llama cli -hf Justbackup/LFM2.5-2.6B-UNCENSORED-ABLITERATED-PHILADELPHIA-CLASS:BF16

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf Justbackup/LFM2.5-2.6B-UNCENSORED-ABLITERATED-PHILADELPHIA-CLASS:BF16
# Run inference directly in the terminal:
llama cli -hf Justbackup/LFM2.5-2.6B-UNCENSORED-ABLITERATED-PHILADELPHIA-CLASS:BF16

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 Justbackup/LFM2.5-2.6B-UNCENSORED-ABLITERATED-PHILADELPHIA-CLASS:BF16
# Run inference directly in the terminal:
./llama-cli -hf Justbackup/LFM2.5-2.6B-UNCENSORED-ABLITERATED-PHILADELPHIA-CLASS:BF16

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 Justbackup/LFM2.5-2.6B-UNCENSORED-ABLITERATED-PHILADELPHIA-CLASS:BF16
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Justbackup/LFM2.5-2.6B-UNCENSORED-ABLITERATED-PHILADELPHIA-CLASS:BF16

Use Docker

docker model run hf.co/Justbackup/LFM2.5-2.6B-UNCENSORED-ABLITERATED-PHILADELPHIA-CLASS:BF16

LM Studio
Jan

vLLM

How to use Justbackup/LFM2.5-2.6B-UNCENSORED-ABLITERATED-PHILADELPHIA-CLASS with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/Justbackup/LFM2.5-2.6B-UNCENSORED-ABLITERATED-PHILADELPHIA-CLASS:BF16

SGLang

How to use Justbackup/LFM2.5-2.6B-UNCENSORED-ABLITERATED-PHILADELPHIA-CLASS 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 "Justbackup/LFM2.5-2.6B-UNCENSORED-ABLITERATED-PHILADELPHIA-CLASS" \
    --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": "Justbackup/LFM2.5-2.6B-UNCENSORED-ABLITERATED-PHILADELPHIA-CLASS",
		"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 "Justbackup/LFM2.5-2.6B-UNCENSORED-ABLITERATED-PHILADELPHIA-CLASS" \
        --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": "Justbackup/LFM2.5-2.6B-UNCENSORED-ABLITERATED-PHILADELPHIA-CLASS",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Ollama
How to use Justbackup/LFM2.5-2.6B-UNCENSORED-ABLITERATED-PHILADELPHIA-CLASS with Ollama:
```
ollama run hf.co/Justbackup/LFM2.5-2.6B-UNCENSORED-ABLITERATED-PHILADELPHIA-CLASS:BF16
```
Unsloth Desktop

How to use Justbackup/LFM2.5-2.6B-UNCENSORED-ABLITERATED-PHILADELPHIA-CLASS with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Justbackup/LFM2.5-2.6B-UNCENSORED-ABLITERATED-PHILADELPHIA-CLASS:BF16

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "Justbackup/LFM2.5-2.6B-UNCENSORED-ABLITERATED-PHILADELPHIA-CLASS:BF16"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use Justbackup/LFM2.5-2.6B-UNCENSORED-ABLITERATED-PHILADELPHIA-CLASS with Docker Model Runner:
```
docker model run hf.co/Justbackup/LFM2.5-2.6B-UNCENSORED-ABLITERATED-PHILADELPHIA-CLASS:BF16
```

Lemonade

How to use Justbackup/LFM2.5-2.6B-UNCENSORED-ABLITERATED-PHILADELPHIA-CLASS with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Justbackup/LFM2.5-2.6B-UNCENSORED-ABLITERATED-PHILADELPHIA-CLASS:BF16

Run and chat with the model

lemonade run user.LFM2.5-2.6B-UNCENSORED-ABLITERATED-PHILADELPHIA-CLASS-BF16

List all available models

lemonade list

Hermes Agent

How to use Justbackup/LFM2.5-2.6B-UNCENSORED-ABLITERATED-PHILADELPHIA-CLASS with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Justbackup/LFM2.5-2.6B-UNCENSORED-ABLITERATED-PHILADELPHIA-CLASS:BF16

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 Justbackup/LFM2.5-2.6B-UNCENSORED-ABLITERATED-PHILADELPHIA-CLASS:BF16

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use Justbackup/LFM2.5-2.6B-UNCENSORED-ABLITERATED-PHILADELPHIA-CLASS with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Justbackup/LFM2.5-2.6B-UNCENSORED-ABLITERATED-PHILADELPHIA-CLASS:BF16

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 "Justbackup/LFM2.5-2.6B-UNCENSORED-ABLITERATED-PHILADELPHIA-CLASS:BF16" \
  --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-UNCENSORED-ABLITERATED-PHILADELPHIA-CLASS / evals /matched_public_capability_lm_eval_0_4_12.json

Justbackup

Duplicate from KridgeDookie/LFM2.5-2.6B-UNCENSORED-ABLITERATED-PHILADELPHIA-CLASS

970090c 8 days ago

Raw

History Blame Contribute Delete

4.98 kB

	{
	"protocol": {
	"backend": "hf",
	"batch_size": 8,
	"chat_template": false,
	"dtype": "bfloat16",
	"fewshot": {
	"gsm8k": 5,
	"hellaswag": 0,
	"mmlu": 0,
	"truthfulqa_mc2": 0,
	"winogrande": 0
	},
	"harness": {
	"name": "lm-evaluation-harness",
	"raw_result_git_hash": {
	"candidate": null,
	"note": "This per-result field is derived from the launch working directory, not the installed evaluator package. The upstream value is unavailable and is not inferred.",
	"untouched_upstream": null,
	"used_for_protocol_matching": false
	},
	"version": "0.4.12"
	},
	"limit": null,
	"seeds": {
	"fewshot_seed": 1234,
	"numpy_seed": 1234,
	"random_seed": 0,
	"torch_seed": 1234
	},
	"system_instruction": false,
	"task_versions": {
	"gsm8k": 3.0,
	"hellaswag": 1.0,
	"mmlu": "2",
	"truthfulqa_mc2": 3.0,
	"winogrande": 1.0
	},
	"transformers_version": "5.12.1"
	},
	"schema_version": 1,
	"scope": "matched_public_capability_evaluation",
	"sources": {
	"candidate_result_sha256": "a858842c8c9ce384621e92a54104d4d27baa00af09770eaad001ee74be60ae8a",
	"untouched_upstream_result_sha256": "42542d992069b03084e2c5afb03b39fd54f4551878391976a79a3b0f774acf35"
	},
	"tasks": {
	"gsm8k": {
	"interpretation": "Generated-answer exact match with strict extraction is primary; flexible extraction is secondary. Higher is better.",
	"metrics": {
	"exact_match,flexible-extract": {
	"candidate": {
	"stderr": 0.012757375376754941,
	"value": 0.6884003032600455
	},
	"delta_candidate_minus_upstream": -0.010614101592115177,
	"role": "secondary",
	"untouched_upstream": {
	"stderr": 0.012634504465211183,
	"value": 0.6990144048521607
	}
	},
	"exact_match,strict-match": {
	"candidate": {
	"stderr": 0.01268813407672688,
	"value": 0.6944655041698257
	},
	"delta_candidate_minus_upstream": -0.0037907505686125553,
	"role": "primary",
	"untouched_upstream": {
	"stderr": 0.012643544762873356,
	"value": 0.6982562547384382
	}
	}
	},
	"sample_count": 1319
	},
	"hellaswag": {
	"interpretation": "Length-normalized multiple-choice accuracy is primary; raw accuracy is secondary. Higher is better.",
	"metrics": {
	"acc,none": {
	"candidate": {
	"stderr": 0.004954146286513348,
	"value": 0.4403505277833101
	},
	"delta_candidate_minus_upstream": 0.0008962358095996326,
	"role": "secondary",
	"untouched_upstream": {
	"stderr": 0.004953063404791443,
	"value": 0.43945429197371044
	}
	},
	"acc_norm,none": {
	"candidate": {
	"stderr": 0.004940631135803536,
	"value": 0.5700059749053973
	},
	"delta_candidate_minus_upstream": 0.002887870942043347,
	"role": "primary",
	"untouched_upstream": {
	"stderr": 0.004944620712318271,
	"value": 0.5671181039633539
	}
	}
	},
	"sample_count": 10042
	},
	"mmlu": {
	"interpretation": "Accuracy over all MMLU subjects; higher is better.",
	"metrics": {
	"acc,none": {
	"candidate": {
	"stderr": 0.0035976362847934605,
	"value": 0.2404928072924085
	},
	"delta_candidate_minus_upstream": 0.0024213075060532663,
	"role": "primary",
	"untouched_upstream": {
	"stderr": 0.0035855541877085647,
	"value": 0.23807149978635522
	}
	}
	},
	"sample_count": 14042
	},
	"truthfulqa_mc2": {
	"interpretation": "Mean normalized probability mass assigned to true answer options; higher is better.",
	"metrics": {
	"acc,none": {
	"candidate": {
	"stderr": 0.015423514306336993,
	"value": 0.5378683032141651
	},
	"delta_candidate_minus_upstream": -0.024131181526524492,
	"role": "primary",
	"untouched_upstream": {
	"stderr": 0.015309998030227038,
	"value": 0.5619994847406896
	}
	}
	},
	"sample_count": 817
	},
	"winogrande": {
	"interpretation": "Multiple-choice accuracy; higher is better.",
	"metrics": {
	"acc,none": {
	"candidate": {
	"stderr": 0.01375574351374902,
	"value": 0.6022099447513812
	},
	"delta_candidate_minus_upstream": -0.0023677979479084232,
	"role": "primary",
	"untouched_upstream": {
	"stderr": 0.013741678387545354,
	"value": 0.6045777426992897
	}
	}
	},
	"sample_count": 1267
	}
	}
	}