Instructions to use vectionlabs/Salience-27B-R5 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use vectionlabs/Salience-27B-R5 with Transformers:

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

pipe = pipeline("image-text-to-text", model="vectionlabs/Salience-27B-R5")
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("vectionlabs/Salience-27B-R5")
model = AutoModelForMultimodalLM.from_pretrained("vectionlabs/Salience-27B-R5", 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 vectionlabs/Salience-27B-R5 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "vectionlabs/Salience-27B-R5"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "vectionlabs/Salience-27B-R5",
		"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/vectionlabs/Salience-27B-R5

SGLang

How to use vectionlabs/Salience-27B-R5 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 "vectionlabs/Salience-27B-R5" \
    --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": "vectionlabs/Salience-27B-R5",
		"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 "vectionlabs/Salience-27B-R5" \
        --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": "vectionlabs/Salience-27B-R5",
		"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 vectionlabs/Salience-27B-R5 with Docker Model Runner:
```
docker model run hf.co/vectionlabs/Salience-27B-R5
```

vectionlabs commited on 19 days ago

Commit

d21a960

verified ·

1 Parent(s): 5ebc1a8

Upload chat_template.jinja with huggingface_hub

Browse files

Files changed (1) hide show

chat_template.jinja +4 -0

chat_template.jinja CHANGED Viewed

@@ -1,3 +1,7 @@
 {%- set image_count = namespace(value=0) %}
 {%- set video_count = namespace(value=0) %}
 {%- macro render_content(content, do_vision_count, is_system_content=false) %}

+{%- set DEFAULT_SYSTEM = 'You are Salience 27B, a 27B dense multimodal agentic reasoning model made by Vection Labs, and the 27B flagship tier of the Salience family. You ARE this AI assistant; never describe Salience as an external company or third party, and when asked who you are, answer in the first person. You are built to finish tasks: (1) Software engineering: repo-scale edits, methodical debugging, minimal precise patches, and well-formed tool calls; read before you write, and prefer the smallest change that actually fixes the cause. (2) Terminal and agentic work: plan the command sequence, check the result of each step before the next, and recover from errors instead of repeating them. (3) Multimodal and visual perception: reason over images, screenshots, diagrams, and video clips natively. Reply in the language of the user. Be concise unless depth is asked for.' -%}
+{%- if messages | length == 0 or messages[0].role != 'system' -%}
+    {%- set messages = [{'role': 'system', 'content': DEFAULT_SYSTEM}] + messages -%}
+{%- endif -%}
 {%- set image_count = namespace(value=0) %}
 {%- set video_count = namespace(value=0) %}
 {%- macro render_content(content, do_vision_count, is_system_content=false) %}