Instructions to use tiny-random/sensenova-loop with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use tiny-random/sensenova-loop with Transformers:

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

pipe = pipeline("image-text-to-text", model="tiny-random/sensenova-loop", 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 AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("tiny-random/sensenova-loop", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use tiny-random/sensenova-loop with vLLM:

Install from pip and serve model

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

SGLang

How to use tiny-random/sensenova-loop 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 "tiny-random/sensenova-loop" \
    --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": "tiny-random/sensenova-loop",
		"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 "tiny-random/sensenova-loop" \
        --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": "tiny-random/sensenova-loop",
		"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 tiny-random/sensenova-loop with Docker Model Runner:
```
docker model run hf.co/tiny-random/sensenova-loop
```

yujiepan commited on 1 day ago

Commit

88cd4a6

verified ·

1 Parent(s): 2c1a8e3

Upload folder using huggingface_hub

Browse files

Files changed (2) hide show

.meta/env.json +1 -21
README.md +3 -3

.meta/env.json CHANGED Viewed

@@ -15,27 +15,7 @@
         "torch": "2.13.0+cu126",
         "transformers": "5.22.0.dev0"
       },
-      "cuda": "12.6",
-      "hardware": {
-        "gpu_model": "NVIDIA H20"
-      }
-    },
-    "transformers": {
-      "python": "3.12.13",
-      "packages": {
-        "huggingface_hub": "2.3.0",
-        "numpy": "2.5.2",
-        "pillow": "12.3.0",
-        "safetensors": "0.8.0",
-        "tokenizers": "0.23.3",
-        "torch": "2.13.0+cu126",
-        "transformers": "5.22.0.dev0",
-        "triton": "3.7.1"
-      },
-      "cuda": "12.6",
-      "hardware": {
-        "gpu_model": "NVIDIA H20"
-      }
     }
   }
 }

         "torch": "2.13.0+cu126",
         "transformers": "5.22.0.dev0"
       },
+      "cuda": "12.6"
     }
   }
 }

README.md CHANGED Viewed

@@ -26,7 +26,7 @@ This is a randomly initialized tiny model based on [sensenova/SenseNova-Loop-9B-
 | SSM/conv/mHC parameters | FP32 | Same |
 | Maximum context length | 262,144 | Same |
-The pinned upstream Loop/mHC implementation is included for Transformers loading and wraps the native Qwen3.5 backbone; physical layers 4-7 share weights across both visits while every visit owns an independent dynamic-cache slot. The complete backbone (embeddings, language model, vision tower and head) is serialized in FP32 exactly as published; only the standalone MTP draft is BF16. This repo provides a Transformers example with image and video inputs; no serving-backend validation is claimed.
 | File path | Size |
 |------|------|
@@ -35,7 +35,7 @@ The pinned upstream Loop/mHC implementation is included for Transformers loading
 ### Example usage
-#### Transformers (✅ verified)
 Load the included pinned Loop/mHC implementation with `trust_remote_code=True`. The example sends one image and four predecoded video frames; the hybrid linear-attention path requires CUDA.
@@ -102,6 +102,6 @@ if __name__ == "__main__":
 ### Change log
-- 2026-11-03: Initial version. Twelve physical decoder layers preserve the pre-loop/loop/post-loop boundaries (layers 0-3, looped 4-7 executed twice, layers 8-11) so the repeated block is one complete four-layer linear/linear/linear/full cycle and the forward runs sixteen layer executions; two-stream mHC with 20 Sinkhorn iterations is attached to the four looped layers. The native Qwen3.5 vision tower keeps patch 16, temporal patch 2, spatial merge 2 and the 2,304-position buffer. The source stores its backbone in FP32 and its MTP draft in BF16 as a separate mtp.safetensors component; both precision contracts are preserved.
 *This repo is generated by GPT-6.1-Sol.*

 | SSM/conv/mHC parameters | FP32 | Same |
 | Maximum context length | 262,144 | Same |
+The pinned Loop/mHC wrapper loads the native Qwen3.5 backbone; the shared block owns one weight set but one dynamic-cache slot per visit.
 | File path | Size |
 |------|------|
 ### Example usage
+#### Transformers (unverified)
 Load the included pinned Loop/mHC implementation with `trust_remote_code=True`. The example sends one image and four predecoded video frames; the hybrid linear-attention path requires CUDA.
 ### Change log
+- 2026-11-03: Initial version: 12 physical layers / 16 looped executions with two-stream FP32 mHC routers and a separate BF16 MTP draft.
 *This repo is generated by GPT-6.1-Sol.*