Instructions to use lloorree/mythxl-70b-gptq with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use lloorree/mythxl-70b-gptq with Transformers:

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

pipe = pipeline("text-generation", model="lloorree/mythxl-70b-gptq")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("lloorree/mythxl-70b-gptq")
model = AutoModelForCausalLM.from_pretrained("lloorree/mythxl-70b-gptq", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use lloorree/mythxl-70b-gptq with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "lloorree/mythxl-70b-gptq"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "lloorree/mythxl-70b-gptq",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/lloorree/mythxl-70b-gptq

SGLang

How to use lloorree/mythxl-70b-gptq 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 "lloorree/mythxl-70b-gptq" \
    --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": "lloorree/mythxl-70b-gptq",
		"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 "lloorree/mythxl-70b-gptq" \
        --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": "lloorree/mythxl-70b-gptq",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use lloorree/mythxl-70b-gptq with Docker Model Runner:
```
docker model run hf.co/lloorree/mythxl-70b-gptq
```

mythxl-70b-gptq / README.md

lloorree

Update README.md

9b69de4 about 3 years ago

preview code

Raw History Blame Contribute Delete

1.05 kB

metadata

license: cc-by-nc-sa-4.0
datasets:
  - kaiokendev/SuperCOT-dataset

Quantized 70B recreation of MythoMax.

Differences:

Includes a 70B recreation of SuperCOT as in the 1.2 version of Huginn
Anywhere Airoboros is merged in, the 1.4.1 version was used instead of 2.X

Known limitation: it strongly prefers novel format in roleplay, and will revert to it over time regardless of context or conversation history.

License is strictly noncommercial, both to match that of its major dependency Chronos 70B and in its own right.

Prompt Format (Copied from the MythoMax page, not necessarily optimal)

This model primarily uses Alpaca formatting, so for optimal model performance, use:

<System prompt/Character Card>

### Instruction:
Your instruction or question here.
For roleplay purposes, I suggest the following - Write <CHAR NAME>'s next reply in a chat between <YOUR NAME> and <CHAR NAME>. Write a single reply only.

### Response: