Instructions to use archit11/small-function-calling with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use archit11/small-function-calling with Transformers:

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

pipe = pipeline("text-generation", model="archit11/small-function-calling")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("archit11/small-function-calling")
model = AutoModelForCausalLM.from_pretrained("archit11/small-function-calling")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use archit11/small-function-calling with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "archit11/small-function-calling"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "archit11/small-function-calling",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/archit11/small-function-calling

SGLang

How to use archit11/small-function-calling 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 "archit11/small-function-calling" \
    --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": "archit11/small-function-calling",
		"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 "archit11/small-function-calling" \
        --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": "archit11/small-function-calling",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use archit11/small-function-calling with Docker Model Runner:
```
docker model run hf.co/archit11/small-function-calling
```

small-function-calling / README.md

archit11

Update README.md

a691d64 verified over 1 year ago

preview code

raw

history blame contribute delete

1.67 kB

	---
	library_name: transformers
	license: mit
	base_model: nisten/Biggie-SmoLlm-0.15B-Base
	tags:
	- generated_from_trainer
	- function_calling
	- function-calling
	- GGUF
	model-index:
	- name: capybara_finetuned_results
	results: []
	datasets:
	- NousResearch/hermes-function-calling-v1
	pipeline_tag: text2text-generation
	---

	<!-- This model card has been generated automatically according to the information the Trainer had access to. You
	should probably proofread and complete it, then remove this comment. -->

	# capybara_finetuned_results

	This model is a fine-tuned version of [nisten/Biggie-SmoLlm-0.15B-Base](https://huggingface.co/nisten/Biggie-SmoLlm-0.15B-Base) on an unknown dataset.
	It achieves the following results on the evaluation set:
	- Loss: 0.0289

	## Model description

	![video](https://0x0.st/XYF7.mp4)
	<video controls autoplay muted src="https://0x0.st/XYF7.mp4"></video>
	## Intended uses & limitations

	More information needed

	## Training and evaluation data

	More information needed

	## Training procedure

	### Training hyperparameters

	The following hyperparameters were used during training:
	- learning_rate: 0.0002
	- train_batch_size: 8
	- eval_batch_size: 8
	- seed: 42
	- gradient_accumulation_steps: 2
	- total_train_batch_size: 16
	- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
	- lr_scheduler_type: cosine
	- lr_scheduler_warmup_steps: 15
	- training_steps: 300

	### Training results

	\| Training Loss \| Epoch \| Step \| Validation Loss \|
	\|:-------------:\|:------:\|:----:\|:---------------:\|
	\| 0.0284 \| 8.4507 \| 300 \| 0.0289 \|


	### Framework versions

	- Transformers 4.44.2
	- Pytorch 2.4.0
	- Datasets 3.0.0
	- Tokenizers 0.19.1