Instructions to use Zyphra/ZAYA1-VL-8B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Zyphra/ZAYA1-VL-8B with Transformers:

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

pipe = pipeline("image-text-to-text", model="Zyphra/ZAYA1-VL-8B")
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 AutoModelForSeq2SeqLM
model = AutoModelForSeq2SeqLM.from_pretrained("Zyphra/ZAYA1-VL-8B", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use Zyphra/ZAYA1-VL-8B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Zyphra/ZAYA1-VL-8B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Zyphra/ZAYA1-VL-8B",
		"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/Zyphra/ZAYA1-VL-8B

SGLang

How to use Zyphra/ZAYA1-VL-8B 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 "Zyphra/ZAYA1-VL-8B" \
    --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": "Zyphra/ZAYA1-VL-8B",
		"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 "Zyphra/ZAYA1-VL-8B" \
        --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": "Zyphra/ZAYA1-VL-8B",
		"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 Zyphra/ZAYA1-VL-8B with Docker Model Runner:
```
docker model run hf.co/Zyphra/ZAYA1-VL-8B
```

hassansh commited on 15 days ago

Commit

7299986

verified ·

1 Parent(s): 311c5a2

Update README.md

Browse files

Files changed (1) hide show

README.md +7 -19

README.md CHANGED Viewed

@@ -45,25 +45,13 @@ ZAYA1-8B-VL builds upon and uses our [ZAYA1-7B LLM](https://huggingface.co/Zyphr
 ZAYA1-8B-VL is trained only upon open data. Detailed dataset descriptions can be found in the accompanying technical report.
-| Eval | ZAYA1-VL-8B-A1B | MolmoE-8B-A1B | DeepSeek-VL2-16B-A2.4B | InternVL3.5-20B-A4B | InternVL3.5-2B | Qwen3-VL-2B | Qwen2.5-VL-3B | Molmo2-4B | Qwen3-VL-4B | InternVL3.5-4B |
-|---|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:|
-| AI2D (test) | **87.5** | <u>73.6</u> | 79.6 | 85.5 | 78.9 | 77.7 | 79.3 | 85.4 | 84.0 | 82.1 |
-| ChartQA (test) | 82.2 | <u>77.9</u> | 84.6 | **87.0** | 81.6 | 78.7 | 83.2 | 86.1 | 81.8 | 86.4 |
-| DocVQA (test) | 92.5 | <u>77.7</u> | 92.3 | 92.9 | 89.4 | 93.3 | 93.9 | 87.8 | **95.3** | 92.4 |
-| InfoVQA (test) | 74.0 | <u>53.9</u> | 75.8 | 78.1 | 70.8 | 72.4 | 77.1 | 78.6 | **80.3** | 78.0 |
-| TextVQA (val) | <u>74.4</u> | 78.1 | **83.4** | 78.5 | 76.5 | 79.9 | 79.2 | 83.1 | 81.5 | 77.6 |
-| OCRBench | 79.8 | <u>55.0</u> | 83.3 | **86.7** | 83.4 | 84.1 | 82.5 | 62.0 | 84.1 | 82.0 |
-| VQA v2.0 (val) | 80.0 | 82.8 | 83.7 | 78.4 | <u>73.6</u> | 78.8 | 79.6 | **85.3** | 80.7 | 76.4 |
-| MathVista (mini) | 64.0 | <u>39.1</u> | 61.2 | **73.5** | 61.4 | 51.8 | 63.2 | 56.5 | 63.6 | 72.8 |
-| MMMU (val) | 46.0 | -- | 46.0 | **72.6** | 49.9 | <u>40.9</u> | 45.7 | 48.8 | 51.4 | 57.2 |
-| SEED (image) | 72.7 | <u>68.7</u> | 76.8 | 76.8 | 75.2 | 74.8 | 73.4 | **78.0** | 77.3 | 76.3 |
-| Blink (val) | <u>45.9</u> | -- | 53.3 | 58.9 | 51.3 | 53.2 | 48.2 | **63.5** | 63.2 | 58.2 |
-| RealWorldQA | 65.0 | <u>60.4</u> | 70.0 | 71.2 | 61.6 | 66.0 | 65.6 | **73.8** | 71.0 | 67.8 |
-| CountBenchQA | 88.1 | 77.4 | 86.0 | 82.1 | <u>70.0</u> | 87.9 | 77.0 | **91.2** | 87.3 | 82.5 |
-| PixMoCount (test) | 83.1 | 45.2 | 38.6 | 47.3 | <u>32.8</u> | 55.7 | 60.0 | 87.0 | **89.2** | 47.3 |
-| Point-Bench (avg) | 58.0 | 58.0 | -- | -- | -- | 53.5 | <u>48.2</u> | **68.5** | 65.1 | -- |
-| RefCOCO (avg) | 84.3 | -- | <u>42.2</u> | **89.1** | 82.9 | 85.0 | 81.0 | -- | 87.8 | 88.8 |
 ## Quick start

 ZAYA1-8B-VL is trained only upon open data. Detailed dataset descriptions can be found in the accompanying technical report.
+Model	AI2D (test)	ChartQA (test)	DocVQA (test)	InfoVQA (test)	TextVQA (val)	OCRBench	VQA v2.0 (val)	MathVista (mini)	MMMU (val)	SEED (image)	Blink (val)	RealWorldQA	CountBenchQA	PixMoCount (test)	Point-Bench (avg)	RefCOCO (avg)
+ZAYA1-VL-8B-A1B	87.5	82.2	92.5	74	74.4	79.8	80	64	46	72.7	45.9	65	88.1	83.1	58	84.3
+MolmoE-8B-A1B	73.6	77.9	77.7	53.9	78.1	55	82.8	39.1	--	68.7	--	60.4	77.4	45.2	58	--
+InternVL3.5-20B-A4B	85.5	87	92.9	78.1	78.5	86.7	78.4	73.5	72.6	76.8	58.9	71.2	82.1	47.3	--	89.1
+Qwen3.5-2B	78.6	78.4			79	83.1	78.3	52.9	49.2	75.8	61	69	84.2	65.5	40.6	80.1
+Molmo2-4B	85.4	86.1	87.8	78.6	83.1	62	85.3	56.5	48.8	78	63.5	73.8	91.2	87	68.5	--
+Qwen3.5-4B	83.7	82.4			81.1	85.3	80.4	82.3	56.9	76.6	56.8	74.2	84.8	84.2	64.4	87.7
 ## Quick start