Instructions to use Meridian-MRM/Samarkand-253m with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Meridian-MRM/Samarkand-253m with Transformers:

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

pipe = pipeline("text-generation", model="Meridian-MRM/Samarkand-253m")

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Meridian-MRM/Samarkand-253m")
model = AutoModelForCausalLM.from_pretrained("Meridian-MRM/Samarkand-253m", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use Meridian-MRM/Samarkand-253m 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 Meridian-MRM/Samarkand-253m:Q2_K
# Run inference directly in the terminal:
llama cli -hf Meridian-MRM/Samarkand-253m:Q2_K

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf Meridian-MRM/Samarkand-253m:Q2_K
# Run inference directly in the terminal:
llama cli -hf Meridian-MRM/Samarkand-253m:Q2_K

Build from source code

git clone https://github.com/ggml-org/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-app
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama serve -hf Meridian-MRM/Samarkand-253m:Q2_K
# Run inference directly in the terminal:
./build/bin/llama cli -hf Meridian-MRM/Samarkand-253m:Q2_K

Use Docker

docker model run hf.co/Meridian-MRM/Samarkand-253m:Q2_K

Llama new
LM Studio
Jan

vLLM

How to use Meridian-MRM/Samarkand-253m with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Meridian-MRM/Samarkand-253m"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Meridian-MRM/Samarkand-253m",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/Meridian-MRM/Samarkand-253m:Q2_K

SGLang

How to use Meridian-MRM/Samarkand-253m 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 "Meridian-MRM/Samarkand-253m" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Meridian-MRM/Samarkand-253m",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "Meridian-MRM/Samarkand-253m" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Meridian-MRM/Samarkand-253m",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Ollama
How to use Meridian-MRM/Samarkand-253m with Ollama:
```
ollama run hf.co/Meridian-MRM/Samarkand-253m:Q2_K
```
Unsloth Desktop
Docker Model Runner
How to use Meridian-MRM/Samarkand-253m with Docker Model Runner:
```
docker model run hf.co/Meridian-MRM/Samarkand-253m:Q2_K
```

Lemonade

How to use Meridian-MRM/Samarkand-253m with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Meridian-MRM/Samarkand-253m:Q2_K

Run and chat with the model

lemonade run user.Samarkand-253m-Q2_K

List all available models

lemonade list

Atomic Chat

Samarkand-253m / tokenizer_config.json

dragosMediv

Add Samarkand-253M: HF safetensors (Qwen3 layout) + GGUF f16

2f46586 verified 6 days ago

Raw History Blame Contribute Delete

247 Bytes

{
  "tokenizer_class": "PreTrainedTokenizerFast",
  "eos_token": "<|endoftext|>",
  "bos_token": "<|endoftext|>",
  "pad_token": "<|endoftext|>",
  "unk_token": "<|endoftext|>",
  "model_max_length": 2048,
  "clean_up_tokenization_spaces": false
}

	{
	"tokenizer_class": "PreTrainedTokenizerFast",
	"eos_token": "<\|endoftext\|>",
	"bos_token": "<\|endoftext\|>",
	"pad_token": "<\|endoftext\|>",
	"unk_token": "<\|endoftext\|>",
	"model_max_length": 2048,
	"clean_up_tokenization_spaces": false
	}