Instructions to use ViorikaAI-org/CalmaCatCoder with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ViorikaAI-org/CalmaCatCoder with Transformers:

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

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

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("ViorikaAI-org/CalmaCatCoder")
model = AutoModelForCausalLM.from_pretrained("ViorikaAI-org/CalmaCatCoder", 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 ViorikaAI-org/CalmaCatCoder with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/ViorikaAI-org/CalmaCatCoder

SGLang

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

Docker Model Runner
How to use ViorikaAI-org/CalmaCatCoder with Docker Model Runner:
```
docker model run hf.co/ViorikaAI-org/CalmaCatCoder
```

CalmaCatCoder / README.md

Dmitriy-Zemskov

Update README.md

7b92d75 verified 9 days ago

preview code

Raw

History Blame Contribute Delete

2.32 kB

	---
	license: other
	license_name: ccpl-1.0
	license_link: LICENSE
	language:
	- en
	pipeline_tag: text-generation
	tags:
	- code
	- qwen2
	- gguf
	- 280m
	library_name: transformers
	---

	# CalmaCatCoder-280M

	CalmaCatCoder-280M is a compact and ultra-fast 280M language model trained from scratch for Python code generation.

	---

	## ⚡ Specs

	* Architecture: Transformer / Causal LM (Qwen2-like)
	* Parameters: ~280M
	* Language: English (Code)
	* Format: ChatML
	* Context: 4096

	---

	## 📜 License

	Distributed under the CalmaCat Public License (CCPL-1.0). See [LICENSE](./LICENSE) for details.

	---

	<details>
	<summary><b>🇷🇺 Нажмите, чтобы открыть описание на русском языке (Click to expand Russian description)</b></summary>

	<br>

	# CalmaCatCoder-280M

	CalmaCatCoder-280M — компактная и ультрабыстрая языковая модель на 280 млн параметров, обученная с нуля для генерации кода на Python.

	---

	## ⚡ Характеристики

	* Архитектура: Transformer / Causal LM (Qwen2-like)
	* Объём параметров: ~280 млн
	* Основной язык: English (Код)
	* Формат диалога: ChatML
	* Контекст: 4096

	---

	## 📜 Лицензия

	Распространяется под кастомной открытой лицензией CalmaCat Public License (CCPL-1.0).
	Полный текст см. в файле [LICENSE](./LICENSE).

	</details>

	---

	## 🚀 Quick Start / Быстрый запуск

	```python
	import torch
	from transformers import AutoModelForCausalLM, AutoTokenizer

	model_id = "ViorikaAI-org/CalmaCatCoder"

	tokenizer = AutoTokenizer.from_pretrained(model_id)
	model = AutoModelForCausalLM.from_pretrained(
	model_id,
	torch_dtype=torch.float16,
	device_map="auto"
	)

	prompt = "<\|im_start\|>user\nWrite a Python function for binary search.<\|im_end\|>\n<\|im_start\|>assistant\n"

	inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
	outputs = model.generate(
	**inputs,
	max_new_tokens=256,
	temperature=0.6,
	top_p=0.9,
	repetition_penalty=1.25,
	do_sample=True
	)

	print(tokenizer.decode(outputs[0], skip_special_tokens=False))
	```