Instructions to use ybelkada/falcon-7b-sharded-bf16 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ybelkada/falcon-7b-sharded-bf16 with Transformers:

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

pipe = pipeline("text-generation", model="ybelkada/falcon-7b-sharded-bf16")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("ybelkada/falcon-7b-sharded-bf16")
model = AutoModelForCausalLM.from_pretrained("ybelkada/falcon-7b-sharded-bf16")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use ybelkada/falcon-7b-sharded-bf16 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ybelkada/falcon-7b-sharded-bf16"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ybelkada/falcon-7b-sharded-bf16",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/ybelkada/falcon-7b-sharded-bf16

SGLang

How to use ybelkada/falcon-7b-sharded-bf16 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 "ybelkada/falcon-7b-sharded-bf16" \
    --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": "ybelkada/falcon-7b-sharded-bf16",
		"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 "ybelkada/falcon-7b-sharded-bf16" \
        --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": "ybelkada/falcon-7b-sharded-bf16",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use ybelkada/falcon-7b-sharded-bf16 with Docker Model Runner:
```
docker model run hf.co/ybelkada/falcon-7b-sharded-bf16
```

Update config.json

by avnishkr - opened Jul 13, 2023

base: refs/heads/main

←

from: refs/pr/7

Discussion Files changed

+11

-10

avnishkr

Jul 13, 2023

No description provided.

Update config.jsone444b198

avnishkr

Jul 13, 2023

•

edited Jul 13, 2023

Original Code is working guys the tiiuae/falcon got the config file restored

##Note: For time being anyone can use the following code while loading:

config = AutoConfig.from_pretrained('https://github.com/avnCode/Falcon-QAMaster/blob/main/config.json')
model = AutoModel.from_pretrained('ybelkada/falcon-7b-sharded-bf16', config=config)
or
model = AutoModelForCausalLM.from_pretrained(
"ybelkada/falcon-7b-sharded-bf16'",
config=config,
return_dict = True,
quantization_config = bnb_config,
device_map = 'auto',
trust_remote_code = True,
)
##

Due to changes in the names of tiiuae/falcon-7b files. Error is popping.
Following are the major changes(for all changes see the Files changed section):
configuration_RW-->configuration_falcon.FalconConfig
modelling_RW-->modeling_falcon.FalconModel
New Function names: FalconConfig, FalconModel, FalconForCausalLM, FalconForQuestionAnswering, FalconForSequenceClassification, FalconForTokenClassification

Please update the config file as modified by me( New config file is tried and tested and its Working!!!)

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

Tap or paste here to upload images

Cannot merge

This branch has merge conflicts in the following files:

config.json

· Sign up or log in to comment