Instructions to use LLM-OS-Models/LFM2.5-8B-A1B-Terminal-ToolBench-Full-SFT-1Epoch with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use LLM-OS-Models/LFM2.5-8B-A1B-Terminal-ToolBench-Full-SFT-1Epoch with Transformers:

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

pipe = pipeline("text-generation", model="LLM-OS-Models/LFM2.5-8B-A1B-Terminal-ToolBench-Full-SFT-1Epoch")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("LLM-OS-Models/LFM2.5-8B-A1B-Terminal-ToolBench-Full-SFT-1Epoch")
model = AutoModelForCausalLM.from_pretrained("LLM-OS-Models/LFM2.5-8B-A1B-Terminal-ToolBench-Full-SFT-1Epoch")
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 LLM-OS-Models/LFM2.5-8B-A1B-Terminal-ToolBench-Full-SFT-1Epoch with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "LLM-OS-Models/LFM2.5-8B-A1B-Terminal-ToolBench-Full-SFT-1Epoch"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "LLM-OS-Models/LFM2.5-8B-A1B-Terminal-ToolBench-Full-SFT-1Epoch",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/LLM-OS-Models/LFM2.5-8B-A1B-Terminal-ToolBench-Full-SFT-1Epoch

SGLang

How to use LLM-OS-Models/LFM2.5-8B-A1B-Terminal-ToolBench-Full-SFT-1Epoch 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 "LLM-OS-Models/LFM2.5-8B-A1B-Terminal-ToolBench-Full-SFT-1Epoch" \
    --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": "LLM-OS-Models/LFM2.5-8B-A1B-Terminal-ToolBench-Full-SFT-1Epoch",
		"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 "LLM-OS-Models/LFM2.5-8B-A1B-Terminal-ToolBench-Full-SFT-1Epoch" \
        --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": "LLM-OS-Models/LFM2.5-8B-A1B-Terminal-ToolBench-Full-SFT-1Epoch",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use LLM-OS-Models/LFM2.5-8B-A1B-Terminal-ToolBench-Full-SFT-1Epoch with Docker Model Runner:
```
docker model run hf.co/LLM-OS-Models/LFM2.5-8B-A1B-Terminal-ToolBench-Full-SFT-1Epoch
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

LLM-OS-Models/LFM2.5-8B-A1B-Terminal-ToolBench-Full-SFT-1Epoch

터미널 작업 자동화를 위한 Terminal SFT 모델입니다. 입력된 작업/이전 터미널 상태를 보고 다음에 실행할 명령을 JSON 형태로 생성하는 용도로 학습했습니다.

모델 요약

Base model: LiquidAI/LFM2.5-8B-A1B
Training setup: full SFT, 1 epoch, Terminal + ToolBench conversation data
Model card snapshot: 2026-06-04 23:09:57 UTC
Corrected TB2-lite score: 52.30
Rank at publication: #1 in the local Terminal README ranking

Quickstart

설치와 로그인:

pip install -U vllm transformers huggingface_hub
huggingface-cli login

관련 코드:

GitHub: https://github.com/LLM-OS-Models/Terminal
vLLM 평가 실행: tb2_lite/scripts/replay_eval.py
chat template/fallback 생성: tb2_lite/scripts/prompt_builder.py
JSON/command 채점: tb2_lite/scripts/replay_metrics.py

vLLM 직접 실행 예시. 평가 코드와 동일하게 chat template을 우선 사용하고, template이 없으면 ChatML/Gemma fallback을 사용합니다.

from transformers import AutoTokenizer
from vllm import LLM, SamplingParams

model_id = "LLM-OS-Models/LFM2.5-8B-A1B-Terminal-ToolBench-Full-SFT-1Epoch"
tp = 1

tokenizer = AutoTokenizer.from_pretrained(model_id, trust_remote_code=True)
llm = LLM(
    model=model_id,
    tokenizer=model_id,
    trust_remote_code=True,
    dtype="bfloat16",
    tensor_parallel_size=tp,
    max_model_len=49152,
    gpu_memory_utilization=0.92,
)

messages = [
    {"role": "system", "content": "You are a terminal automation assistant. Return JSON only."},
    {"role": "user", "content": "Inspect the current directory and list Python files."},
]

def render_chatml(messages):
    parts = []
    for message in messages:
        role = "assistant" if message["role"] == "assistant" else message["role"]
        if role == "tool":
            role = "user"
        parts.append(f"<|im_start|>{role}\n{message['content']}<|im_end|>\n")
    parts.append("<|im_start|>assistant\n")
    return "".join(parts)

def render_gemma4_turn(messages, empty_thought_channel=False):
    parts = ["<bos>"]
    for message in messages:
        role = "model" if message["role"] == "assistant" else message["role"]
        if role == "tool":
            role = "user"
        parts.append(f"<|turn>{role}\n{message['content'].strip()}<turn|>\n")
    parts.append("<|turn>model\n")
    if empty_thought_channel:
        parts.append("<|channel>thought\n<channel|>")
    return "".join(parts)

def render_prompt(model_id, tokenizer, messages):
    model_key = model_id.lower()
    if "gemma-4" in model_key:
        try:
            return tokenizer.apply_chat_template(
                messages,
                tokenize=False,
                add_generation_prompt=True,
                enable_thinking=False,
            )
        except Exception:
            return render_gemma4_turn(
                messages,
                empty_thought_channel=("26b" in model_key or "31b" in model_key),
            )
    try:
        return tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
    except Exception:
        return render_chatml(messages)

prompt = render_prompt(model_id, tokenizer, messages)
sampling = SamplingParams(
    temperature=0.0,
    top_p=1.0,
    max_tokens=1024,
    repetition_penalty=1.0,
)
outputs = llm.generate([prompt], sampling_params=sampling)
print(outputs[0].outputs[0].text)

권장 출력 형식:

{
  "analysis": "brief reasoning about the next terminal action",
  "plan": "short execution plan",
  "commands": [
    {"keystrokes": "ls -la\n", "duration": 0.1}
  ],
  "task_complete": false
}

평가와 동일한 replay 명령:

python tb2_lite/scripts/replay_eval.py \
  --model LLM-OS-Models/LFM2.5-8B-A1B-Terminal-ToolBench-Full-SFT-1Epoch \
  --model-short LLM-OS-Models__LFM2.5-8B-A1B-Terminal-ToolBench-Full-SFT-1Epoch \
  --eval-path tb2_lite/data/replay_full.jsonl \
  --output-dir /home/work/.data/tb2_lite_eval/corrected_readme_models_vllm \
  --dtype bfloat16 \
  --tp 1 \
  --max-model-len 49152 \
  --max-tokens 1024 \
  --temperature 0.0 \
  --top-p 1.0 \
  --gpu-memory-utilization 0.92 \
  --language-model-only

기본 권장 tensor parallel: 1. OOM이면 --tp와 tensor_parallel_size를 2/4/8로 올리세요.
corrected TB2-lite 평가는 temperature=0.0, top_p=1.0, max_tokens=1024로 고정했습니다.
Gemma 4는 JSON 출력을 위해 enable_thinking=False를 사용하고, 26B/31B 계열은 평가 코드에서 empty thought channel 처리를 자동 적용합니다.

평가 상태

Corrected TB2-lite score: 52.30
Cmd F1: 0.5230
Precision: 0.5854
Recall: 0.5431
First command exact: 49.5%
Valid JSON: 76.9%
Sec/Step: 0.087
Evaluation split: corrected TB2-lite full replay, 303/303 steps
Result JSON: tb2_lite/results/20260605T_live_hrm_lora_lfm_epoch1/LFM2.5-8B-A1B-terminal-toolbench-full-1epoch-checkpoint-1542.json

모델군 해석

이 1epoch checkpoint는 corrected TB2-lite 기준 Score 52.30으로 업로드 시점의 Terminal README 전체 1위입니다.
같은 run의 2epoch checkpoint Score 50.48보다 +1.82 높았습니다. 2epoch는 command coverage가 약간 줄고 JSON 안정성도 조금 낮아져, 이 데이터 구성에서는 1epoch가 더 좋은 선택입니다.
강점은 command coverage입니다. Recall 0.5431과 Precision 0.5854가 같이 높아, 기존 30~40점대 모델보다 정답 command set을 더 넓게 복원합니다.
약점은 완전한 JSON 안정성입니다. Valid JSON 76.9%라 strong baseline이지만, invalid JSON과 empty command case는 후속 JSON-only target 정제로 더 줄일 수 있습니다.
TB2-lite 점수는 일반 지능 벤치마크가 아니라 터미널 next-action JSON 재현 능력을 측정합니다.
생성 명령은 실제 실행 전에 sandbox, allowlist, human review 같은 안전장치를 거쳐야 합니다.