Text Generation
Transformers
Safetensors
English
llama
tinystories
language-model
educational
text-generation-inference
Instructions to use manojredhat/tiny-llama with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use manojredhat/tiny-llama with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="manojredhat/tiny-llama")# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("manojredhat/tiny-llama") model = AutoModelForCausalLM.from_pretrained("manojredhat/tiny-llama") - Notebooks
- Google Colab
- Kaggle
- Local Apps
- vLLM
How to use manojredhat/tiny-llama with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "manojredhat/tiny-llama" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "manojredhat/tiny-llama", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/manojredhat/tiny-llama
- SGLang
How to use manojredhat/tiny-llama 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 "manojredhat/tiny-llama" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "manojredhat/tiny-llama", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'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 "manojredhat/tiny-llama" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "manojredhat/tiny-llama", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use manojredhat/tiny-llama with Docker Model Runner:
docker model run hf.co/manojredhat/tiny-llama
File size: 622 Bytes
82332a9 8f4b37b 756e974 82332a9 756e974 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 | {
"architectures": [
"LlamaForCausalLM"
],
"attention_dropout": 0.0,
"bos_token_id": 1,
"eos_token_id": 2,
"hidden_act": "silu",
"hidden_size": 288,
"initializer_range": 0.02,
"intermediate_size": 768,
"max_position_embeddings": 256,
"model_type": "llama",
"num_attention_heads": 6,
"num_hidden_layers": 6,
"num_key_value_heads": 6,
"pad_token_id": 0,
"pretraining_tp": 1,
"rms_norm_eps": 1e-05,
"rope_scaling": null,
"rope_theta": 10000.0,
"tie_word_embeddings": false,
"torch_dtype": "float32",
"transformers_version": "4.36.0",
"use_cache": true,
"vocab_size": 512
}
|