Instructions to use teeem92/alina-voice-7b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use teeem92/alina-voice-7b with Transformers:

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

pipe = pipeline("text-generation", model="teeem92/alina-voice-7b")
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("teeem92/alina-voice-7b")
model = AutoModelForCausalLM.from_pretrained("teeem92/alina-voice-7b", 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]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use teeem92/alina-voice-7b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "teeem92/alina-voice-7b"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "teeem92/alina-voice-7b",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/teeem92/alina-voice-7b

SGLang

How to use teeem92/alina-voice-7b 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 "teeem92/alina-voice-7b" \
    --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": "teeem92/alina-voice-7b",
		"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 "teeem92/alina-voice-7b" \
        --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": "teeem92/alina-voice-7b",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use teeem92/alina-voice-7b with Docker Model Runner:
```
docker model run hf.co/teeem92/alina-voice-7b
```

Alina Voice Bot 7B (LPTracker & ElevenLabs)

Автономная языковая модель голосового телефонного робота Алины (компания LPTracker), дообученная из Qwen/Qwen2.5-7B-Instruct с помощью Soup CLI (Layer Streaming).

Особенности и правила синтеза (33 правила LPTracker):

Синтез ElevenLabs: оптимизирована для генерации живой речи без монотонности.
Акустические теги: встроены метки пауз и естественных вздохов (... [hmm] ..., ... [sigh] ..., ... [chuckle] ...).
Лимит длины: реплики оптимизированы до 70 символов для минимизации задержек в телефонии.
Числа словами: числа пишутся прописью для корректного чтения голосовым движком.
Никаких восклицательных знаков: устраняет искажения и неестественные вскрики синтезатора.
Специфика: женский род («поняла», «услышала»), мессенджер «вацаап».

Использование через Transformers:

from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "teeem92/alina-voice-7b"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto", torch_dtype="auto")

messages = [
    {"role": "system", "content": "Ты голосовой робот Алина компании LPTracker..."},
    {"role": "user", "content": "Здравствуйте, а вы кто такие?"}
]
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
out = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(out[0][inputs.input_ids.shape[1]:], skip_special_tokens=True))

Downloads last month: 480

Safetensors

Model size

8B params

Tensor type

F16

Model tree for teeem92/alina-voice-7b

Base model

Qwen/Qwen2.5-7B

Finetuned

Qwen/Qwen2.5-7B-Instruct

Finetuned

(3250)

this model