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

Libraries

How to use monotykamary/LFM2.5-2.6B-RLCD with Transformers:

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

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

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("monotykamary/LFM2.5-2.6B-RLCD")
model = AutoModelForCausalLM.from_pretrained("monotykamary/LFM2.5-2.6B-RLCD", 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 monotykamary/LFM2.5-2.6B-RLCD with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/monotykamary/LFM2.5-2.6B-RLCD

SGLang

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

Docker Model Runner
How to use monotykamary/LFM2.5-2.6B-RLCD with Docker Model Runner:
```
docker model run hf.co/monotykamary/LFM2.5-2.6B-RLCD
```

LFM2.5-2.6B-RLCD / ARTIFACTS.json

monotykamary

feat: add verified inference-only parallel constrained decoding

3145545 verified 9 days ago

Raw

History Blame Contribute Delete

2.72 kB

	{
	"files": {
	"requirements-pcd.txt": "3aa238d3352bda6f720b8c48c3f7cbeb6ddad4bac4fe1d94bf27924d077e1d83",
	"lfm25_pcd_modal.py": "67454084f7a646509b707034e4bfe0f0ed60ae23cac1539e0baa7967d1408dbc",
	"README-PCD.md": "67543cb872a0b40757e081e830be423a323fe1637e0896b5b1ade72fcc57c2b6",
	"LICENSE-CODE": "6c6d531bb55d1105ce57394979dbc2dc6cdeb0f9f5ef39856489378a71f380b0",
	"THIRD_PARTY_NOTICES.md": "f38a5df33965de672354ff43efb691f3d726d4378ac71ff84a183ca43234243a",
	"ruff.toml": "d5eaf462a77af92f39be2cf5e345936451eb717403a8b380cb918e8b5440fdf1",
	"pcd/__init__.py": "691b9d06e8effb0595a1a5c63906624bcdc6b4c2d482f1fc0f674d52439cad72",
	"pcd/benchmark.py": "98cda0b4bfa688391cdf2a492443831473c80a5f3e2f1b14ccf9eb5f1cee9bf5",
	"pcd/cache.py": "213061dc97d9a3b3315d7063868a12b1ce4dce167ef7d9cdaa84980ae08fb6bb",
	"pcd/config.py": "18f6a7fa6b84cbd7c48aeb2e60db1e2cb056a28d0bb406ad3d8e1c02a2c7bbc9",
	"pcd/engine.py": "f369d8ade685d45318cfd703e1d683d75749f27e1b73d4ec4c04ed3e65825036",
	"pcd/prompting.py": "a4220d122a2e8c10c687d37d2a525e32959b2c3928020fac8e90d55af94be7bb",
	"pcd/publishing.py": "2a299e45367e760b4d67b42b706fb200ec18e793bb295cfd652910b458bf24ae",
	"pcd/schema.py": "b70db7a1d5bb5ce41df485142167a1802711d267a46ef46ccd1cf254baf0ebfd",
	"pcd/tasks.py": "b76ca325125fc2d4304d1e395a81e32caa02165c1ad8451e1f5d7a73742db968",
	"tests/pcd/conftest.py": "39c365aed040ea13bad97f11998ccaae860a0b2750e57de0d3e11284a6deadd9",
	"tests/pcd/test_engine.py": "a059463869bc12f9abbaad3a51b7be741f0905dc54d1174f73189c6e2889e324",
	"tests/pcd/test_release.py": "7b6f82c2e9b702e7d576f4238a9655dcc1b088e467f6ec17a1c015b1217c8a20",
	"scripts/build_pcd_release.py": "1290c88e7e63f40b2e4e2b45a88d60618a8f132cbcf5e65e19a6966fdbab6431",
	"scripts/report_pcd.py": "5ceda179cd8715b4504fda93523c4b4037a907c72b2d0c0f04a6536bcc0cc507",
	"docs/pcd-results.md": "476944292e7b4c4b57e57f35e8998d9b7d4ca2d35d5031778ed4b85bcff46083",
	"results/pcd/20260916-175747-benchmark-diagnostic-1f57e2.json": "5d2a22097d19d537fe0cd847a53ab5d563e43d2287050d08241d063c7d5efeaa",
	"results/pcd/20260916-175954-benchmark-stress-d12f1b.json": "5f6b37889fdb92567d14a681fd63fc0137e248f9ec5ba37ad7cc8322fdd3961f",
	"results/pcd/20260916-180625-validate-diagnostic-0d3ba0.json": "9f460555954cb6facb43894d9a44b732fd7e1bf504e12c552740975462f29ba5"
	},
	"reports": [
	"results/pcd/20260916-175747-benchmark-diagnostic-1f57e2.json",
	"results/pcd/20260916-175954-benchmark-stress-d12f1b.json",
	"results/pcd/20260916-180625-validate-diagnostic-0d3ba0.json"
	],
	"core_files": [
	"engine.py",
	"cache.py",
	"schema.py",
	"prompting.py",
	"config.py",
	"tasks.py"
	]
	}