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

Libraries

How to use Azazelle/smol_bruin-7b with Transformers:

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

pipe = pipeline("text-generation", model="Azazelle/smol_bruin-7b")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Azazelle/smol_bruin-7b")
model = AutoModelForCausalLM.from_pretrained("Azazelle/smol_bruin-7b")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use Azazelle/smol_bruin-7b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Azazelle/smol_bruin-7b"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Azazelle/smol_bruin-7b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/Azazelle/smol_bruin-7b

SGLang

How to use Azazelle/smol_bruin-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 "Azazelle/smol_bruin-7b" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Azazelle/smol_bruin-7b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "Azazelle/smol_bruin-7b" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Azazelle/smol_bruin-7b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use Azazelle/smol_bruin-7b with Docker Model Runner:
```
docker model run hf.co/Azazelle/smol_bruin-7b
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Model Card for smol_bruin-7b

Slerp merge of go-bruins-v2 and smol-7b.

.yaml file for mergekit

slices:
  - sources:
      - model: rwitz/go-bruins-v2
        layer_range: [0, 32]
      - model: rishiraj/smol-7b
        layer_range: [0, 32]
merge_method: slerp
base_model: mistralai/Mistral-7B-v0.1
parameters:
  t:
    - filter: self_attn
      value: [0.44, 0.72, 0.61, 0.83, 1]
    - filter: mlp
      value: [0.56, 0.28, 0.39, 0.17, 0]
    - value: 0.5 # fallback for rest of tensors
dtype: float16

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	71.05
AI2 Reasoning Challenge (25-Shot)	67.58
HellaSwag (10-Shot)	86.48
MMLU (5-Shot)	65.05
TruthfulQA (0-shot)	55.65
Winogrande (5-shot)	81.14
GSM8k (5-shot)	70.43

Downloads last month: 236

Safetensors

Model size

7B params

Tensor type

F16

Model tree for Azazelle/smol_bruin-7b

Quantizations

3 models

Evaluation results

normalized accuracy on AI2 Reasoning Challenge (25-Shot)
test set Open LLM Leaderboard

67.580
normalized accuracy on HellaSwag (10-Shot)
validation set Open LLM Leaderboard

86.480
accuracy on MMLU (5-Shot)
test set Open LLM Leaderboard

65.050
mc2 on TruthfulQA (0-shot)
validation set Open LLM Leaderboard

55.650
accuracy on Winogrande (5-shot)
validation set Open LLM Leaderboard

81.140
accuracy on GSM8k (5-shot)
test set Open LLM Leaderboard

70.430