Instructions to use agentionai/Signal-3.8-27B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use agentionai/Signal-3.8-27B with Transformers:

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

pipe = pipeline("image-text-to-text", model="agentionai/Signal-3.8-27B")
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("agentionai/Signal-3.8-27B")
model = AutoModelForMultimodalLM.from_pretrained("agentionai/Signal-3.8-27B", device_map="auto")
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 agentionai/Signal-3.8-27B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "agentionai/Signal-3.8-27B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "agentionai/Signal-3.8-27B",
		"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/agentionai/Signal-3.8-27B

SGLang

How to use agentionai/Signal-3.8-27B 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 "agentionai/Signal-3.8-27B" \
    --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": "agentionai/Signal-3.8-27B",
		"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 "agentionai/Signal-3.8-27B" \
        --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": "agentionai/Signal-3.8-27B",
		"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 agentionai/Signal-3.8-27B with Docker Model Runner:
```
docker model run hf.co/agentionai/Signal-3.8-27B
```

Loop occurred after like 10M tokens :)

by JC1DA - opened 10 days ago

Discussion

JC1DA

10 days ago

Thanks for the model. It seems great for short-medium running task.
I tried to build a game (candy crush saga with some custom configurations) from scratch and it started to loop after around 15-30 mins

agentionai

Owner 10 days ago

Thanks, that is useful. I would like to reproduce it. Could you share:

which file (tier) and runtime, and the sampling settings (temperature, top-p, top-k, min-p, repeat penalty)
the context size the server was started with, and whether the loop began around the time the context filled (llama.cpp logs "context shift" when it does)
which agent or harness drove it, and whether the MTP draft head was on

Two things to try meanwhile. Run with the sampling from the card (temperature 0.7, top-p 0.95, top-k 20, min-p 0, repeat penalty 1.0); greedy and temperature 1.0 both degrade this model on long generations. And give the agent a context of 64k or more with -c 65536; a 30-minute build overflows 32k, and once the window shifts the model loses the start of its own code.

We will run the same kind of task here, a game built from scratch, Signal against the base model, and report what we find.

JC1DA

9 days ago

•

edited 9 days ago

let me try with temperature 0.7 again. I was using sglang with default suggested temperature 1.0 as in the original model with full context window.

agentionai

Owner 9 days ago

I might upload a milder version

Shockem

9 days ago

I also noticed hallucinations without temperature being set to 0.7, though setting it to 0.7 fixed it on my end.

agentionai

Owner 9 days ago

I have replicated looping and will upload new files today

agentionai changed discussion status to closed 9 days ago

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