Instructions to use trl-internal-testing/tiny-LlavaForConditionalGeneration with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use trl-internal-testing/tiny-LlavaForConditionalGeneration with Transformers:

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

pipe = pipeline("image-text-to-text", model="trl-internal-testing/tiny-LlavaForConditionalGeneration")
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 AutoProcessor, AutoModelForImageTextToText

processor = AutoProcessor.from_pretrained("trl-internal-testing/tiny-LlavaForConditionalGeneration")
model = AutoModelForImageTextToText.from_pretrained("trl-internal-testing/tiny-LlavaForConditionalGeneration")
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?"}
        ]
    },
]
inputs = processor.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use trl-internal-testing/tiny-LlavaForConditionalGeneration with vLLM:

Install from pip and serve model

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

SGLang

How to use trl-internal-testing/tiny-LlavaForConditionalGeneration 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 "trl-internal-testing/tiny-LlavaForConditionalGeneration" \
    --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": "trl-internal-testing/tiny-LlavaForConditionalGeneration",
		"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 "trl-internal-testing/tiny-LlavaForConditionalGeneration" \
        --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": "trl-internal-testing/tiny-LlavaForConditionalGeneration",
		"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 trl-internal-testing/tiny-LlavaForConditionalGeneration with Docker Model Runner:
```
docker model run hf.co/trl-internal-testing/tiny-LlavaForConditionalGeneration
```

qgallouedec HF Staff commited on Jul 20, 2025

Commit

1c6c707

verified ·

1 Parent(s): 919d519

Upload LlavaForConditionalGeneration

Browse files

Files changed (3) hide show

config.json +20 -3
generation_config.json +1 -1
model.safetensors +2 -2

config.json CHANGED Viewed

@@ -2,29 +2,46 @@
   "architectures": [
     "LlavaForConditionalGeneration"
   ],
-  "ignore_index": -100,
   "image_seq_length": 576,
   "image_token_index": 32000,
   "model_type": "llava",
   "projector_hidden_act": "gelu",
   "text_config": {
     "head_dim": 2,
     "hidden_size": 8,
     "intermediate_size": 32,
     "model_type": "llama",
     "num_attention_heads": 4,
     "num_hidden_layers": 2,
     "num_key_value_heads": 2,
     "vocab_size": 32005
   },
   "torch_dtype": "float32",
-  "transformers_version": "4.47.0.dev0",
   "vision_config": {
-    "hidden_size": 8,
     "image_size": 336,
     "intermediate_size": 32,
     "model_type": "clip_vision_model",
     "num_attention_heads": 4,
     "num_hidden_layers": 2,
     "patch_size": 14,
     "projection_dim": 8

   "architectures": [
     "LlavaForConditionalGeneration"
   ],
   "image_seq_length": 576,
   "image_token_index": 32000,
   "model_type": "llava",
+  "multimodal_projector_bias": true,
   "projector_hidden_act": "gelu",
   "text_config": {
+    "attention_bias": false,
+    "attention_dropout": 0.0,
     "head_dim": 2,
+    "hidden_act": "silu",
     "hidden_size": 8,
+    "initializer_range": 0.02,
     "intermediate_size": 32,
+    "max_position_embeddings": 2048,
+    "mlp_bias": false,
     "model_type": "llama",
     "num_attention_heads": 4,
     "num_hidden_layers": 2,
     "num_key_value_heads": 2,
+    "pretraining_tp": 1,
+    "rms_norm_eps": 1e-06,
+    "rope_scaling": null,
+    "rope_theta": 10000.0,
+    "use_cache": true,
     "vocab_size": 32005
   },
   "torch_dtype": "float32",
+  "transformers_version": "4.54.0.dev0",
   "vision_config": {
+    "attention_dropout": 0.0,
+    "hidden_act": "quick_gelu",
+    "hidden_size": 16,
     "image_size": 336,
+    "initializer_factor": 1.0,
+    "initializer_range": 0.02,
     "intermediate_size": 32,
+    "layer_norm_eps": 1e-05,
     "model_type": "clip_vision_model",
     "num_attention_heads": 4,
+    "num_channels": 3,
     "num_hidden_layers": 2,
     "patch_size": 14,
     "projection_dim": 8

generation_config.json CHANGED Viewed

@@ -2,5 +2,5 @@
   "_from_model_config": true,
   "bos_token_id": 1,
   "eos_token_id": 2,
-  "transformers_version": "4.47.0.dev0"
 }

   "_from_model_config": true,
   "bos_token_id": 1,
   "eos_token_id": 2,
+  "transformers_version": "4.54.0.dev0"
 }

model.safetensors CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:eb60b1bc3c41b94aec9c4c24111b3dd3a5b1853317fce7d967cbea4263464e1a
-size 2108896

 version https://git-lfs.github.com/spec/v1
+oid sha256:5f54b0ec090dd9d4ec1955951dc2d1449bcf1cffebcee2f651554858609f23c6
+size 2157456