Instructions to use hustvl/mmMamba-linear with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use hustvl/mmMamba-linear with Transformers:

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

pipe = pipeline("image-text-to-text", model="hustvl/mmMamba-linear", 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)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("hustvl/mmMamba-linear", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use hustvl/mmMamba-linear with vLLM:

Install from pip and serve model

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

SGLang

How to use hustvl/mmMamba-linear 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 "hustvl/mmMamba-linear" \
    --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": "hustvl/mmMamba-linear",
		"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 "hustvl/mmMamba-linear" \
        --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": "hustvl/mmMamba-linear",
		"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 hustvl/mmMamba-linear with Docker Model Runner:
```
docker model run hf.co/hustvl/mmMamba-linear
```

HongyuanTao commited on Feb 26, 2025

Commit

975bc02

verified ·

1 Parent(s): 300e9bb

Update modeling_mmMamba_embedding.py

Browse files

Files changed (1) hide show

modeling_mmMamba_embedding.py +4 -2

modeling_mmMamba_embedding.py CHANGED Viewed

@@ -410,7 +410,7 @@ class MHA_LM(nn.Module):
         ):
             if self.rotary_emb_dim > 0:
                 q, kv = self.rotary_emb(
-                    q, kv, seqlen_offset=seqlen_offset, max_seqlen=rotary_max_seqlen
                 )
             if inference_params is None:
                 k, v = kv.unbind(dim=-3)
@@ -538,7 +538,9 @@ class Mamba2_LM(nn.Module):
         conv_state, ssm_state = None, None
         if inference_params is not None:
             conv_state, ssm_state = self._get_states_from_cache(inference_params, batch)
         if use_cache and inference_params.seqlen_offset==0:
             vkq, new_conv_states = causal_conv1d_fn(
                 vkq.transpose(1, 2),

         ):
             if self.rotary_emb_dim > 0:
                 q, kv = self.rotary_emb(
+                    q, kv, seqlen_offset=seqlen_offset[:bsz,...], max_seqlen=rotary_max_seqlen
                 )
             if inference_params is None:
                 k, v = kv.unbind(dim=-3)
         conv_state, ssm_state = None, None
         if inference_params is not None:
             conv_state, ssm_state = self._get_states_from_cache(inference_params, batch)
+        conv_state = conv_state[:batch, ...]
+        ssm_state = ssm_state[:batch, ...]
         if use_cache and inference_params.seqlen_offset==0:
             vkq, new_conv_states = causal_conv1d_fn(
                 vkq.transpose(1, 2),