Instructions to use InternScience/Agents-A1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use InternScience/Agents-A1 with Transformers:

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

pipe = pipeline("text-generation", model="InternScience/Agents-A1")
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 AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("InternScience/Agents-A1")
model = AutoModelForMultimodalLM.from_pretrained("InternScience/Agents-A1")
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?"}
        ]
    },
]
inputs = processor.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use InternScience/Agents-A1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "InternScience/Agents-A1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "InternScience/Agents-A1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/InternScience/Agents-A1

SGLang

How to use InternScience/Agents-A1 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 "InternScience/Agents-A1" \
    --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": "InternScience/Agents-A1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "InternScience/Agents-A1" \
        --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": "InternScience/Agents-A1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use InternScience/Agents-A1 with Docker Model Runner:
```
docker model run hf.co/InternScience/Agents-A1
```

Stability Comparison: Agents-A1 vs Nex-N2-mini under Q5_K_M Quantization

#12

by BipedalBit - opened 1 day ago

Discussion

BipedalBit

1 day ago

•

edited 1 day ago

When using the Q5_K_M quantized version (LordNeel/Agents-A1-GGUF, bartowski/nex-agi_Nex-N2-mini-GGUF) for both models with the same llama.cpp settings (-ctk q4_0 -ctv q4_0 --ctx-size 196608) and the chat_template from froggeric/Qwen-Fixed-Chat-Templates, I observed that in the hermes-agent conversation mode for complex troubleshooting scenarios, the stability of Agents-A1 during long-horizon multi-turn tool calls is inferior to that of Nex-N2-mini. And I've switched my primary hermes-agent model back to Nex-N2-mini.
Sorry, I can't share the actual conversation logs that show the difference—they're tightly coupled with my private deployment and ops setup.

sY713

Intern Science org 1 day ago

When using the Q5_K_M quantized version (LordNeel/Agents-A1-GGUF, bartowski/nex-agi_Nex-N2-mini-GGUF) for both models with the same llama.cpp settings (-ctk q4_0 -ctv q4_0 --ctx-size 196608) and the chat_template from froggeric/Qwen-Fixed-Chat-Templates, I observed that in the hermes-agent conversation mode for complex troubleshooting scenarios, the stability of Agents-A1 during long-horizon multi-turn tool calls is inferior to that of Nex-N2-mini. And I've switched my primary hermes-agent model back to Nex-N2-mini.
Sorry, I can't share the actual conversation logs that show the difference—they're tightly coupled with my private deployment and ops setup.

Thanks a lot for your sharing! We're really happy to hear about your real-world feedback and from other community devs too, and we totally get that you can't share full conversations due to privacy concerns. If possible, though, we'd love it if you could share some of your task descriptions, or point out which specific tools had failed calls. That would really help us deliver better open-source models for the community!

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