Instructions to use WithinUsAI/Llama3.2-Hermes.Dolphin.Coder-1B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use WithinUsAI/Llama3.2-Hermes.Dolphin.Coder-1B with Transformers:

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

pipe = pipeline("text-generation", model="WithinUsAI/Llama3.2-Hermes.Dolphin.Coder-1B")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("WithinUsAI/Llama3.2-Hermes.Dolphin.Coder-1B")
model = AutoModelForCausalLM.from_pretrained("WithinUsAI/Llama3.2-Hermes.Dolphin.Coder-1B")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use WithinUsAI/Llama3.2-Hermes.Dolphin.Coder-1B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "WithinUsAI/Llama3.2-Hermes.Dolphin.Coder-1B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "WithinUsAI/Llama3.2-Hermes.Dolphin.Coder-1B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/WithinUsAI/Llama3.2-Hermes.Dolphin.Coder-1B

SGLang

How to use WithinUsAI/Llama3.2-Hermes.Dolphin.Coder-1B 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 "WithinUsAI/Llama3.2-Hermes.Dolphin.Coder-1B" \
    --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": "WithinUsAI/Llama3.2-Hermes.Dolphin.Coder-1B",
		"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 "WithinUsAI/Llama3.2-Hermes.Dolphin.Coder-1B" \
        --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": "WithinUsAI/Llama3.2-Hermes.Dolphin.Coder-1B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use WithinUsAI/Llama3.2-Hermes.Dolphin.Coder-1B with Docker Model Runner:
```
docker model run hf.co/WithinUsAI/Llama3.2-Hermes.Dolphin.Coder-1B
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Llama-3.2-HermesDolphin-Coder-1B

Llama-3.2-HermesDolphin-Coder-1B is a compact merged language model designed for general instruction following, coding assistance, and lightweight conversational use. It combines Hermes-style instruction tuning and Dolphin-style helpfulness into a small Llama 3.2 class model intended for experimentation, local workflows, and developer-oriented prompting.

This repository appears to be a merge model created with mergekit using the SLERP merge method.

Model Summary

Model type: Causal language model
Architecture: LlamaForCausalLM
Primary use: Text generation, instruction following, code-oriented prompting
Library: Transformers
Merge method: SLERP
Format: Safetensors

Base Models

This merged model is based on:

artificialguybr/LLAMA-3.2-1B-OpenHermes2.5
dphn/Dolphin3.0-Llama3.2-1B
meta-llama/Llama-3.2-1B-Instruct

Merge Details

According to the repository metadata/configuration, the merge was produced with mergekit using a SLERP setup with a midpoint interpolation parameter.

Merge configuration

merge_method: slerp
base_model: artificialguybr/LLAMA-3.2-1B-OpenHermes2.5
models:
  - model: dphn/Dolphin3.0-Llama3.2-1B
    parameters:
      weight: 1.0
dtype: float32
parameters:
  t: 0.5