Instructions to use Agnes-AI/Agnes-3.0-Flash with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Agnes-AI/Agnes-3.0-Flash with Transformers:

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

pipe = pipeline("image-text-to-text", model="Agnes-AI/Agnes-3.0-Flash", trust_remote_code=True)
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("Agnes-AI/Agnes-3.0-Flash", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Agnes-AI/Agnes-3.0-Flash with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Agnes-AI/Agnes-3.0-Flash"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Agnes-AI/Agnes-3.0-Flash",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/Agnes-AI/Agnes-3.0-Flash

SGLang

How to use Agnes-AI/Agnes-3.0-Flash 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 "Agnes-AI/Agnes-3.0-Flash" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Agnes-AI/Agnes-3.0-Flash",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

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 "Agnes-AI/Agnes-3.0-Flash" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Agnes-AI/Agnes-3.0-Flash",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Docker Model Runner
How to use Agnes-AI/Agnes-3.0-Flash with Docker Model Runner:
```
docker model run hf.co/Agnes-AI/Agnes-3.0-Flash
```

Please remove the misleading benchmarks

by linuxid10t - opened 4 days ago

Discussion

linuxid10t

4 days ago

If you are going to publish a "preview" checkpoint, do not use the benchmarks from the full, finished version. This is highly misleading. Better yet, just publish the weights from the release version. I understand that is a business decision, but it would be far better for the community to just publish the real deal and keep the pro model proprietary.

aquilesfd

3 days ago

the benchmarks are for the preview version it seems, but it is noticeably worse than the final checkpoint

GPQA Diamond: 85.1 vs 92.4
SciCode: 38.1 vs 51.6
AA-Omniscience: 23 vs 25
AA-LCR: 68.3 vs 81

and $0.005/0.05/0.15 is crazy economics for a 30B dense model. i don't think Agnes 3.0 Flash and Flash Preview are the same model.

l3afai

2 days ago

its a dense?? i thought it was a MoE

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