Instructions to use tiny-random/llama-4-8E with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use tiny-random/llama-4-8E with Transformers:

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

pipe = pipeline("text-generation", model="tiny-random/llama-4-8E")
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("tiny-random/llama-4-8E")
model = AutoModelForImageTextToText.from_pretrained("tiny-random/llama-4-8E")
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 tiny-random/llama-4-8E with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "tiny-random/llama-4-8E"
# 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/llama-4-8E",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/tiny-random/llama-4-8E

SGLang

How to use tiny-random/llama-4-8E 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/llama-4-8E" \
    --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/llama-4-8E",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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/llama-4-8E" \
        --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/llama-4-8E",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use tiny-random/llama-4-8E with Docker Model Runner:
```
docker model run hf.co/tiny-random/llama-4-8E
```

yujiepan commited on Apr 10, 2025

Commit

a8c0b5b

verified ·

1 Parent(s): 0da80d0

Upload folder using huggingface_hub

Browse files

Files changed (2) hide show

special_tokens_map.json +1 -1
tokenizer_config.json +1 -1

special_tokens_map.json CHANGED Viewed

@@ -14,7 +14,7 @@
     "single_word": false
   },
   "pad_token": {
-    "content": "<|finetune_right_pad_id|>",
     "lstrip": false,
     "normalized": false,
     "rstrip": false,

     "single_word": false
   },
   "pad_token": {
+    "content": "<|finetune_right_pad|>",
     "lstrip": false,
     "normalized": false,
     "rstrip": false,

tokenizer_config.json CHANGED Viewed

@@ -9091,7 +9091,7 @@
     "attention_mask"
   ],
   "model_max_length": 1048576,
-  "pad_token": "<|finetune_right_pad_id|>",
   "processor_class": "Llama4Processor",
   "tokenizer_class": "PreTrainedTokenizer"
 }

     "attention_mask"
   ],
   "model_max_length": 1048576,
+  "pad_token": "<|finetune_right_pad|>",
   "processor_class": "Llama4Processor",
   "tokenizer_class": "PreTrainedTokenizer"
 }