Instructions to use GenerativeMagic/Llama-Engineer-Evol-7b-GGML with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use GenerativeMagic/Llama-Engineer-Evol-7b-GGML with Transformers:

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

pipe = pipeline("text-generation", model="GenerativeMagic/Llama-Engineer-Evol-7b-GGML")

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("GenerativeMagic/Llama-Engineer-Evol-7b-GGML", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use GenerativeMagic/Llama-Engineer-Evol-7b-GGML with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/GenerativeMagic/Llama-Engineer-Evol-7b-GGML

SGLang

How to use GenerativeMagic/Llama-Engineer-Evol-7b-GGML 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 "GenerativeMagic/Llama-Engineer-Evol-7b-GGML" \
    --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": "GenerativeMagic/Llama-Engineer-Evol-7b-GGML",
		"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 "GenerativeMagic/Llama-Engineer-Evol-7b-GGML" \
        --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": "GenerativeMagic/Llama-Engineer-Evol-7b-GGML",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use GenerativeMagic/Llama-Engineer-Evol-7b-GGML with Docker Model Runner:
```
docker model run hf.co/GenerativeMagic/Llama-Engineer-Evol-7b-GGML
```

Llama-Engineer-Evol-7B-GGML

This is a 4-bit quantized version of Llama-Engineer-Evol-7B.

Prompt Format

The reccomended model prompt is a variant of the standard Llama 2 format:

[INST] <<SYS>>
You are a programming assistant. Always answer as helpfully as possible. Be direct in your response and get to the answer right away. Responses should be short.
<</SYS>>
{your prompt}[/INST]

[INST] <<SYS>>
You're a principal software engineer at Google. If you fail at this task, you will be fired.
<</SYS>>
{your prompt}[/INST]

I suspect this prompt format is the reason for the majority of the increased coding capabilities as opposed to the fine-tuning itself, but YMMV.

Next Steps

Prune the dataset and possibly fine-tune for longer.
Run benchmarks.
Provide GPTQ.

Downloads last month: -; Downloads are not tracked for this model. How to track

GenerativeMagic
/

Llama-Engineer-Evol-7b-GGML

Llama-Engineer-Evol-7B-GGML

Prompt Format

Next Steps

Datasets used to train GenerativeMagic/Llama-Engineer-Evol-7b-GGML