Instructions to use HuggingFaceM4/idefics-80b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use HuggingFaceM4/idefics-80b with Transformers:

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

pipe = pipeline("text-generation", model="HuggingFaceM4/idefics-80b")

# Load model directly
from transformers import AutoProcessor, AutoModelForImageTextToText

processor = AutoProcessor.from_pretrained("HuggingFaceM4/idefics-80b")
model = AutoModelForImageTextToText.from_pretrained("HuggingFaceM4/idefics-80b")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use HuggingFaceM4/idefics-80b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "HuggingFaceM4/idefics-80b"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "HuggingFaceM4/idefics-80b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/HuggingFaceM4/idefics-80b

SGLang

How to use HuggingFaceM4/idefics-80b 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 "HuggingFaceM4/idefics-80b" \
    --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": "HuggingFaceM4/idefics-80b",
		"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 "HuggingFaceM4/idefics-80b" \
        --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": "HuggingFaceM4/idefics-80b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use HuggingFaceM4/idefics-80b with Docker Model Runner:
```
docker model run hf.co/HuggingFaceM4/idefics-80b
```

sasha commited on Aug 28, 2023

Commit

a2b0422

1 Parent(s): 4e645cd

Adding carbon footprint information

Browse files

Calculated the carbon footprint of different stages of IDEFICS training + experimentation based on information provided by @VictorSanh and @stas !

Files changed (1) hide show

README.md +26 -0

README.md CHANGED Viewed

@@ -317,6 +317,32 @@ The IDEFICS models were trained on an AWS SageMaker cluster with 8x80GB A100 GPU
 The training software is built on top of HuggingFace Transformers + Accelerate, and [DeepSpeed ZeRO-3](https://github.com/microsoft/DeepSpeed) for training, and [WebDataset](https://github.com/webdataset/webdataset) for data loading.
 # Bias, Risks, and Limitations
 Significant research has explored bias and fairness issues with language models (see, e.g., [Sheng et al. (2021)](https://aclanthology.org/2021.acl-long.330.pdf) and [Bender et al. (2021)](https://dl.acm.org/doi/pdf/10.1145/3442188.3445922)).

 The training software is built on top of HuggingFace Transformers + Accelerate, and [DeepSpeed ZeRO-3](https://github.com/microsoft/DeepSpeed) for training, and [WebDataset](https://github.com/webdataset/webdataset) for data loading.
+## Environmental Impact
+We distinguish the 3 phases of the creation of IDEFICS and report our carbon emissions separately for each one of them:
+*Preliminary experimentation*
+- **Hardware Type:** Intel Cascade Lake CPUs, NVIDIA V100 and A100 GPUs
+- **Hours used:** 460,000 CPU hours, 385,000 V100 GPU hours, and 300,000 A100 GPU hours
+- **Cloud Provider:** N/A (Jean Zay cluster)
+- **Compute Region:** France (57g CO2eq/kWh)
+- **Carbon Emitted:** 16,714 kgs of CO2eq
+*IDEFICS-80B pretraining*
+- **Hardware Type:** 512 NVIDIA A100 GPUs
+- **Hours used:** 672 hours (28 days)
+- **Cloud Provider:** AWS
+- **Compute Region:** US-West 2 (288g CO2eq/kWh)
+- **Carbon Emitted:** 39,498 kgs of CO2eq
+*IDEFICS-80b-instruct finetuning*
+- **Hardware Type:** 384 NVIDIA A100 GPUs
+- **Hours used:** 72 hours (3 days)
+- **Cloud Provider:** AWS
+- **Compute Region:** US-West 2 (288g CO2eq/kWh)
+- **Carbon Emitted:** 3,174 kgs of CO2eq
 # Bias, Risks, and Limitations
 Significant research has explored bias and fairness issues with language models (see, e.g., [Sheng et al. (2021)](https://aclanthology.org/2021.acl-long.330.pdf) and [Bender et al. (2021)](https://dl.acm.org/doi/pdf/10.1145/3442188.3445922)).