Instructions to use RekaAI/reka-edge-2603 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use RekaAI/reka-edge-2603 with Transformers:

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

pipe = pipeline("image-text-to-text", model="RekaAI/reka-edge-2603", 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)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoModelForImageTextToText
model = AutoModelForImageTextToText.from_pretrained("RekaAI/reka-edge-2603", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use RekaAI/reka-edge-2603 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "RekaAI/reka-edge-2603"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "RekaAI/reka-edge-2603",
		"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/RekaAI/reka-edge-2603

SGLang

How to use RekaAI/reka-edge-2603 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 "RekaAI/reka-edge-2603" \
    --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": "RekaAI/reka-edge-2603",
		"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 "RekaAI/reka-edge-2603" \
        --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": "RekaAI/reka-edge-2603",
		"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 RekaAI/reka-edge-2603 with Docker Model Runner:
```
docker model run hf.co/RekaAI/reka-edge-2603
```

kwajiehao commited on Apr 8

Commit

7876ea0

verified ·

1 Parent(s): a67e042

Upload convert_reka_vlm_to_gguf.py with huggingface_hub

Browse files

Files changed (1) hide show

convert_reka_vlm_to_gguf.py +8 -1

convert_reka_vlm_to_gguf.py CHANGED Viewed

@@ -284,7 +284,14 @@ class RekaYasa2VisionMmprojModel(base.MmprojModel):
     def tensor_force_quant(
         self, name: str, new_name: str, bid: int | None, n_dims: int
     ) -> base.gguf.GGMLQuantizationType | bool:
-        del name, new_name, bid
         if n_dims > 1:
             return base.gguf.GGMLQuantizationType.F16
         return False

     def tensor_force_quant(
         self, name: str, new_name: str, bid: int | None, n_dims: int
     ) -> base.gguf.GGMLQuantizationType | bool:
+        del name, bid
+        if new_name in {
+            "v.patch_embd.weight",
+            "v.vision_pos_embed",
+        }:
+            return base.gguf.GGMLQuantizationType.F32
         if n_dims > 1:
             return base.gguf.GGMLQuantizationType.F16
         return False