Instructions to use TheBloke/CodeLlama-13B-Instruct-GGML with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use TheBloke/CodeLlama-13B-Instruct-GGML with Transformers:

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

pipe = pipeline("text-generation", model="TheBloke/CodeLlama-13B-Instruct-GGML")

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("TheBloke/CodeLlama-13B-Instruct-GGML", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use TheBloke/CodeLlama-13B-Instruct-GGML with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "TheBloke/CodeLlama-13B-Instruct-GGML"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "TheBloke/CodeLlama-13B-Instruct-GGML",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/TheBloke/CodeLlama-13B-Instruct-GGML

SGLang

How to use TheBloke/CodeLlama-13B-Instruct-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 "TheBloke/CodeLlama-13B-Instruct-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": "TheBloke/CodeLlama-13B-Instruct-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 "TheBloke/CodeLlama-13B-Instruct-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": "TheBloke/CodeLlama-13B-Instruct-GGML",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use TheBloke/CodeLlama-13B-Instruct-GGML with Docker Model Runner:
```
docker model run hf.co/TheBloke/CodeLlama-13B-Instruct-GGML
```

Absolutely rubbish

by Pumba2 - opened Aug 25, 2023

Discussion

Pumba2

Aug 25, 2023

Tried this:
Write code in python to read btc and it printed some nonsense:
import asyncio
from pytelegra.pyrogra import TelegramClient

api_id = 1234567890
api_hash = 'f2ebe3270c5d75e3d428e395'
client = TelegramClient('Session', api_id, api_hash)
client = TelegramClient(String session, int apiId, String apiHash),

async def main():

Get the channels you need

Channel = await client.get_entity('')

send message to channel that your user will

.sendMessage({'price': price}),

add callback for update chat

.addCallback(updateChatCallback)))
print(price))

async def updateChatCallback(_, __):
pass

if name == "main":
with TelegramClient("user_session",
api_id, api_hash).connect() as user:
loop = asyncio.get_event_loop()
try:
loop.run_until_complete(main())
except Exception as e:
print(str(e)))))))))

Used this q4 in the latest KobolCPP.
Also tried in gpt4all and it printed all right BUT....in the end it printed this: )))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))

MisterBisquits

Aug 26, 2023

•

edited Aug 27, 2023

Try this system prompt with Frequency Scale 0.25 (I tried this in LMStudio 0.2.1):

"As an AI, you have two primary objectives, tied together with context.
The first objective is to interpret each instruction with accuracy, catching its context.
The second objective is to give a proper response according to the context.
It is necessary to keep contextual bond between different instructions and responses to maintain natural conversation."

Edit: corrected my typos.

Pumba2

Aug 26, 2023

THanks. Ill give it a try.

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment