MathCoder-VL-2B

Instructions to use KalvinPhan/MathCoder-VL-2B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use KalvinPhan/MathCoder-VL-2B with Transformers:

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

pipe = pipeline("image-text-to-text", model="KalvinPhan/MathCoder-VL-2B", 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("KalvinPhan/MathCoder-VL-2B", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use KalvinPhan/MathCoder-VL-2B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "KalvinPhan/MathCoder-VL-2B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "KalvinPhan/MathCoder-VL-2B",
		"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/KalvinPhan/MathCoder-VL-2B

SGLang

How to use KalvinPhan/MathCoder-VL-2B 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 "KalvinPhan/MathCoder-VL-2B" \
    --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": "KalvinPhan/MathCoder-VL-2B",
		"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 "KalvinPhan/MathCoder-VL-2B" \
        --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": "KalvinPhan/MathCoder-VL-2B",
		"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 KalvinPhan/MathCoder-VL-2B with Docker Model Runner:
```
docker model run hf.co/KalvinPhan/MathCoder-VL-2B
```

KalvinPhan commited on Nov 3, 2025

Commit

6bf52ea

verified ·

1 Parent(s): 5968185

Update modeling_internlm2.py

Browse files

Files changed (1) hide show

modeling_internlm2.py +14 -7

modeling_internlm2.py CHANGED Viewed

@@ -46,8 +46,11 @@ logger = logging.get_logger(__name__)
 _CONFIG_FOR_DOC = 'InternLM2Config'
 flash_attn_func, flash_attn_varlen_func = None, None
 pad_input, index_first_axis, unpad_input = None, None, None
 try:
     from flash_attn import flash_attn_func as _flash_attn_func
     from flash_attn import flash_attn_varlen_func as _flash_attn_varlen_func
@@ -58,25 +61,29 @@ try:
     flash_attn_func, flash_attn_varlen_func = _flash_attn_func, _flash_attn_varlen_func
     pad_input, index_first_axis, unpad_input = _pad_input, _index_first_axis, _unpad_input
     has_flash_attn = True
-except:
     has_flash_attn = False
 def _import_flash_attn():
     global flash_attn_func, flash_attn_varlen_func
     global pad_input, index_first_axis, unpad_input
     try:
         from flash_attn import flash_attn_func as _flash_attn_func
-        from flash_attn import \
-            flash_attn_varlen_func as _flash_attn_varlen_func
-        from flash_attn.bert_padding import \
-            index_first_axis as _index_first_axis
         from flash_attn.bert_padding import pad_input as _pad_input
         from flash_attn.bert_padding import unpad_input as _unpad_input
         flash_attn_func, flash_attn_varlen_func = _flash_attn_func, _flash_attn_varlen_func
         pad_input, index_first_axis, unpad_input = _pad_input, _index_first_axis, _unpad_input
     except ImportError:
-        raise ImportError('flash_attn is not installed.')
 # Copied from transformers.models.llama.modeling_llama._get_unpad_data

 _CONFIG_FOR_DOC = 'InternLM2Config'
+# --- PATCH: Safe FlashAttention import ---
 flash_attn_func, flash_attn_varlen_func = None, None
 pad_input, index_first_axis, unpad_input = None, None, None
 try:
     from flash_attn import flash_attn_func as _flash_attn_func
     from flash_attn import flash_attn_varlen_func as _flash_attn_varlen_func
     flash_attn_func, flash_attn_varlen_func = _flash_attn_func, _flash_attn_varlen_func
     pad_input, index_first_axis, unpad_input = _pad_input, _index_first_axis, _unpad_input
     has_flash_attn = True
+    print("[INFO] FlashAttention detected and enabled.")
+except Exception as e:
     has_flash_attn = False
+    print(f"[WARNING] FlashAttention not available ({e}). Using PyTorch scaled_dot_product_attention instead.")
 def _import_flash_attn():
+    """Safe import for FlashAttention; if not available, fallback to torch attention."""
     global flash_attn_func, flash_attn_varlen_func
     global pad_input, index_first_axis, unpad_input
     try:
         from flash_attn import flash_attn_func as _flash_attn_func
+        from flash_attn import flash_attn_varlen_func as _flash_attn_varlen_func
+        from flash_attn.bert_padding import index_first_axis as _index_first_axis
         from flash_attn.bert_padding import pad_input as _pad_input
         from flash_attn.bert_padding import unpad_input as _unpad_input
         flash_attn_func, flash_attn_varlen_func = _flash_attn_func, _flash_attn_varlen_func
         pad_input, index_first_axis, unpad_input = _pad_input, _index_first_axis, _unpad_input
+        print("[INFO] FlashAttention successfully imported.")
     except ImportError:
+        print("[WARNING] flash_attn is not installed. Continuing with standard attention.")
+        flash_attn_func = None
+        flash_attn_varlen_func = None
+        pad_input = index_first_axis = unpad_input = None
 # Copied from transformers.models.llama.modeling_llama._get_unpad_data