Instructions to use bruhzair/prototype-0.4x311 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use bruhzair/prototype-0.4x311 with Transformers:

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

pipe = pipeline("text-generation", model="bruhzair/prototype-0.4x311")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("bruhzair/prototype-0.4x311")
model = AutoModelForCausalLM.from_pretrained("bruhzair/prototype-0.4x311")
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 bruhzair/prototype-0.4x311 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "bruhzair/prototype-0.4x311"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "bruhzair/prototype-0.4x311",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/bruhzair/prototype-0.4x311

SGLang

How to use bruhzair/prototype-0.4x311 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 "bruhzair/prototype-0.4x311" \
    --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": "bruhzair/prototype-0.4x311",
		"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 "bruhzair/prototype-0.4x311" \
        --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": "bruhzair/prototype-0.4x311",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use bruhzair/prototype-0.4x311 with Docker Model Runner:
```
docker model run hf.co/bruhzair/prototype-0.4x311
```

prototype-0.4x311

This is a merge of pre-trained language models created using mergekit.

Merge Details

Merge Method

This model was merged using the DELLA merge method using /workspace/prototype-0.4x295 as a base.

Models Merged

The following models were included in the merge:

/workspace/cache/models--AstroMLab--AstroSage-70B/snapshots/86496984f418ef5a6825f2c16983595e7f7d5930
/workspace/cache/models--shisa-ai--shisa-v2-llama3.3-70b/snapshots/0a3080fbcbfbb0160c30db82b05be039453a4c01
/workspace/cache/models--LumiOpen--Llama-Poro-2-70B-Instruct/snapshots/ba7a467a544e2b8d944a8a8636120fd0fea9358d
/workspace/cache/models--deepcogito--cogito-v2-preview-llama-70B/snapshots/1e1d12e8eaebd6084a8dcf45ecdeaa2f4b8879ce
/workspace/cache/models--TheDrummer--Fallen-Llama-3.3-70B-v1/snapshots/d46ef2629f1c3cd46789a55793c5ff0af60de3e8
/workspace/cache/models--watt-ai--watt-tool-70B/snapshots/dbe19344ec6ee4b9e1636e9e6ce24fc6a85a725e

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: /workspace/cache/models--shisa-ai--shisa-v2-llama3.3-70b/snapshots/0a3080fbcbfbb0160c30db82b05be039453a4c01
    parameters:
      weight: 0.16
      density: 0.7
      epsilon: 0.2
  - model: /workspace/cache/models--AstroMLab--AstroSage-70B/snapshots/86496984f418ef5a6825f2c16983595e7f7d5930
    parameters:
      weight: 0.16
      density: 0.7
      epsilon: 0.2
  - model: /workspace/cache/models--LumiOpen--Llama-Poro-2-70B-Instruct/snapshots/ba7a467a544e2b8d944a8a8636120fd0fea9358d
    parameters:
      weight: 0.16
      density: 0.7
      epsilon: 0.2
  - model: /workspace/cache/models--watt-ai--watt-tool-70B/snapshots/dbe19344ec6ee4b9e1636e9e6ce24fc6a85a725e
    parameters:
      weight: 0.16
      density: 0.7
      epsilon: 0.2
  - model: /workspace/cache/models--TheDrummer--Fallen-Llama-3.3-70B-v1/snapshots/d46ef2629f1c3cd46789a55793c5ff0af60de3e8
    parameters:
      weight: 0.16
      density: 0.7
      epsilon: 0.2
  - model: /workspace/cache/models--deepcogito--cogito-v2-preview-llama-70B/snapshots/1e1d12e8eaebd6084a8dcf45ecdeaa2f4b8879ce
    parameters:
      weight: 0.2
      density: 0.5
      epsilon: 0.25
base_model: /workspace/prototype-0.4x295
merge_method: della
parameters:
  normalize: false
  lambda: 1.05
chat_template: llama3
pad_to_multiple_of: 8
int8_mask: true
tokenizer:
 source: base
dtype: float32

Downloads last month: -

Safetensors

Model size

71B params

Tensor type

F32

Paper for bruhzair/prototype-0.4x311

DELLA-Merging: Reducing Interference in Model Merging through Magnitude-Based Sampling

Paper • 2406.11617 • Published Jun 17, 2024 • 10