Instructions to use MTSAIR/Cotype-Nano with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use MTSAIR/Cotype-Nano with Transformers:

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

pipe = pipeline("text-generation", model="MTSAIR/Cotype-Nano")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

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

tokenizer = AutoTokenizer.from_pretrained("MTSAIR/Cotype-Nano")
model = AutoModelForCausalLM.from_pretrained("MTSAIR/Cotype-Nano", 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=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Inference
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use MTSAIR/Cotype-Nano with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/MTSAIR/Cotype-Nano

SGLang

How to use MTSAIR/Cotype-Nano 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 "MTSAIR/Cotype-Nano" \
    --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": "MTSAIR/Cotype-Nano",
		"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 "MTSAIR/Cotype-Nano" \
        --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": "MTSAIR/Cotype-Nano",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use MTSAIR/Cotype-Nano with Docker Model Runner:
```
docker model run hf.co/MTSAIR/Cotype-Nano
```

AlanRobotics commited on Nov 22, 2024

Commit

bdd3ce2

verified ·

1 Parent(s): 1a1ed0d

Update README.md

Browse files

Files changed (1) hide show

README.md +4 -4

README.md CHANGED Viewed

@@ -8,9 +8,9 @@ pipeline_tag: text-generation
 # Cotype-Nano🤖
-MTSAIR/Cotype-Nano – это легковесный ИИ на основе meta-llama/Llama-3.2-1B, разработанный для выполнения задач с минимальными ресурсами. Он оптимизирован для быстрого и эффективного взаимодействия с пользователями, обеспечивая высокую производительность даже в условиях ограниченных ресурсов.*
-Cotype Nano is a lightweight AI based on meta-llama/Llama-3.2-1B, designed to perform tasks with minimal resources. It is optimized for fast and efficient interaction with users, providing high performance even under resource-constrained conditions.
 ### Inference with vLLM
 ```sh
@@ -103,11 +103,11 @@ print(res[0]['generated_text'][-1]['content'])
 The model was trained in two stages. In the first stage, MLP layers were trained on mathematics and code. In the second stage, the entire model was trained on internal and open synthetic instructional datasets.
-### ru-llm-arena: **21.3** (local measurement)
 | **Model**                                   | **Score** | **95% CI**      | **Avg Tokens** |
 | ------------------------------------------- | --------- | --------------- | -------------- |
-| **MTSAIR/Cotype-Nano**                  | **29.4**     | **+1.7 / -1.6**     | **616**         |
 | storm-7b                                    | 20.62     | +2.0 / -1.6     | 419.32         |
 | neural-chat-7b-v3-3                         | 19.04     | +2.0 / -1.7     | 927.21         |
 | Vikhrmodels-Vikhr-Llama-3.2-1B-instruct | 19.04 | +1.3 / -1.6 | 958.63     |

 # Cotype-Nano🤖
+MTSAIR/Cotype-Nano – это легковесный ИИ, разработанный для выполнения задач с минимальными ресурсами. Он оптимизирован для быстрого и эффективного взаимодействия с пользователями, обеспечивая высокую производительность даже в условиях ограниченных ресурсов.*
+Cotype Nano is a lightweight AI, designed to perform tasks with minimal resources. It is optimized for fast and efficient interaction with users, providing high performance even under resource-constrained conditions.
 ### Inference with vLLM
 ```sh
 The model was trained in two stages. In the first stage, MLP layers were trained on mathematics and code. In the second stage, the entire model was trained on internal and open synthetic instructional datasets.
+### ru-llm-arena: **29.4** (local measurement)
 | **Model**                                   | **Score** | **95% CI**      | **Avg Tokens** |
 | ------------------------------------------- | --------- | --------------- | -------------- |
+| **MTSAIR/Cotype-Nano**                  | **29.4**     | **+1.7 / -1.6**     | **588**         |
 | storm-7b                                    | 20.62     | +2.0 / -1.6     | 419.32         |
 | neural-chat-7b-v3-3                         | 19.04     | +2.0 / -1.7     | 927.21         |
 | Vikhrmodels-Vikhr-Llama-3.2-1B-instruct | 19.04 | +1.3 / -1.6 | 958.63     |