Instructions to use TuralBayev/axeron-forge-34431f2a with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use TuralBayev/axeron-forge-34431f2a with Transformers:

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

pipe = pipeline("text-generation", model="TuralBayev/axeron-forge-34431f2a")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("TuralBayev/axeron-forge-34431f2a")
model = AutoModelForCausalLM.from_pretrained("TuralBayev/axeron-forge-34431f2a", 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 TuralBayev/axeron-forge-34431f2a with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "TuralBayev/axeron-forge-34431f2a"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "TuralBayev/axeron-forge-34431f2a",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/TuralBayev/axeron-forge-34431f2a

SGLang

How to use TuralBayev/axeron-forge-34431f2a 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 "TuralBayev/axeron-forge-34431f2a" \
    --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": "TuralBayev/axeron-forge-34431f2a",
		"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 "TuralBayev/axeron-forge-34431f2a" \
        --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": "TuralBayev/axeron-forge-34431f2a",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use TuralBayev/axeron-forge-34431f2a with Docker Model Runner:
```
docker model run hf.co/TuralBayev/axeron-forge-34431f2a
```

axeron-forge-34431f2a / README.md

Tural Bayev

Upload folder using huggingface_hub

e3a6220 verified about 2 months ago

preview code

Raw

History Blame Contribute Delete

1.25 kB

	---
	base_model:
	- Qwen/Qwen2.5-Coder-3B-Instruct
	- Qwen/Qwen2.5-3B
	- Qwen/Qwen2.5-3B-Instruct
	library_name: transformers
	tags:
	- forgelm
	- forge

	---
	# 34431f2a-b1de-4115-9c9e-83406cb6bae0

	This is a forge of pre-trained language models created using [forgelm](https://github.com/AxeronAI/forgelm).

	## Forge Details
	### Forge Method

	This model was forged using the [Task Arithmetic](https://arxiv.org/abs/2212.04089) forge method using [Qwen/Qwen2.5-3B](https://huggingface.co/Qwen/Qwen2.5-3B) as a base.

	### Models Forged

	The following models were included in the forge:
	* [Qwen/Qwen2.5-Coder-3B-Instruct](https://huggingface.co/Qwen/Qwen2.5-Coder-3B-Instruct)
	* [Qwen/Qwen2.5-3B-Instruct](https://huggingface.co/Qwen/Qwen2.5-3B-Instruct)

	### Configuration

	The following YAML configuration was used to produce this model:

	```yaml
	base_model: Qwen/Qwen2.5-3B
	dtype: bfloat16
	forge_method: task_arithmetic
	modules:
	default:
	slices:
	- sources:
	- layer_range: [0, 36]
	model: Qwen/Qwen2.5-3B-Instruct
	parameters:
	weight: 0.5
	- layer_range: [0, 36]
	model: Qwen/Qwen2.5-Coder-3B-Instruct
	parameters:
	weight: -0.1
	- layer_range: [0, 36]
	model: Qwen/Qwen2.5-3B
	```