Instructions to use ICBU-NPU/FashionGPT-70B-V1.1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ICBU-NPU/FashionGPT-70B-V1.1 with Transformers:

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

pipe = pipeline("text-generation", model="ICBU-NPU/FashionGPT-70B-V1.1")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("ICBU-NPU/FashionGPT-70B-V1.1")
model = AutoModelForCausalLM.from_pretrained("ICBU-NPU/FashionGPT-70B-V1.1", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use ICBU-NPU/FashionGPT-70B-V1.1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ICBU-NPU/FashionGPT-70B-V1.1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ICBU-NPU/FashionGPT-70B-V1.1",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/ICBU-NPU/FashionGPT-70B-V1.1

SGLang

How to use ICBU-NPU/FashionGPT-70B-V1.1 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 "ICBU-NPU/FashionGPT-70B-V1.1" \
    --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": "ICBU-NPU/FashionGPT-70B-V1.1",
		"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 "ICBU-NPU/FashionGPT-70B-V1.1" \
        --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": "ICBU-NPU/FashionGPT-70B-V1.1",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use ICBU-NPU/FashionGPT-70B-V1.1 with Docker Model Runner:
```
docker model run hf.co/ICBU-NPU/FashionGPT-70B-V1.1
```

GTPQ, GGUF ver request

by Yhyu13 - opened Sep 19, 2023

Discussion

Yhyu13

Sep 19, 2023

@TheBloke
Would you like to make the quantized version of this model for the community. It is current ranked #1 on the LLM leaderboard, and also is the #1 model to surpass 74 avg score

delphijb

Sep 21, 2023

•

edited Sep 21, 2023

Hello,
https://huggingface.co/TheBloke/FashionGPT-70B-V1.1-GGUF

:-)

Edit:
Without holding back enterprising people, or even innovation, I notice that the author doesn't seem very active, or even in a hurry to put information into his models.
==> His first template was created 4 days ago, and is still empty.

This is a problem because we don't know what's inside, or how it was formed (Only FashionGPT ? or a mix with other databases?
For example, if the author includes test data (ARC, HellaSwag, MMLU, TruthfulQA, ...) in his training set, it will logically be stronger, in a biased way of course. Over time, we'll have to see if he stays number 1... or just stays in this ranking.
(I notice that Marcoroni-70b has disappeared. And indeed, I tested it, and it didn't seem any stronger than my Orca_v3_70b).

All right, I'm a pessimist! Let's give him a chance :-)

Yhyu13

Sep 21, 2023

Yep, its done!

Yhyu13 changed discussion status to closed Sep 21, 2023

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