Instructions to use edsoncarvalhointuria/merged-schema-forg-ai with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use edsoncarvalhointuria/merged-schema-forg-ai with Transformers:

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

pipe = pipeline("text-generation", model="edsoncarvalhointuria/merged-schema-forg-ai", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("edsoncarvalhointuria/merged-schema-forg-ai", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("edsoncarvalhointuria/merged-schema-forg-ai", trust_remote_code=True, device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.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=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use edsoncarvalhointuria/merged-schema-forg-ai with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "edsoncarvalhointuria/merged-schema-forg-ai"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "edsoncarvalhointuria/merged-schema-forg-ai",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/edsoncarvalhointuria/merged-schema-forg-ai

SGLang

How to use edsoncarvalhointuria/merged-schema-forg-ai 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 "edsoncarvalhointuria/merged-schema-forg-ai" \
    --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": "edsoncarvalhointuria/merged-schema-forg-ai",
		"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 "edsoncarvalhointuria/merged-schema-forg-ai" \
        --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": "edsoncarvalhointuria/merged-schema-forg-ai",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use edsoncarvalhointuria/merged-schema-forg-ai with Docker Model Runner:
```
docker model run hf.co/edsoncarvalhointuria/merged-schema-forg-ai
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Schema Forg AI

Overview

Schema Forg AI is a specialized language model designed to generate production-ready API schemas using the @asteasolutions/zod-to-openapi library.

Unlike generic code generation models, Schema Forg AI was specifically fine-tuned to understand the relationship between JSON structures, OpenAPI specifications, and Zod schemas, producing code compatible with the zod-to-openapi ecosystem.

The goal of the project is to simplify API documentation workflows by generating a single source of truth based on Zod, from which OpenAPI JSON and YAML specifications can be automatically exported.

What the model can do

The model can generate:

Zod schemas
@asteasolutions/zod-to-openapi code
OpenAPI JSON
OpenAPI YAML

Starting from inputs such as:

Raw JSON
Existing OpenAPI specifications
API endpoint definitions
Structured objects

Why this model?

The JavaScript ecosystem already provides excellent tools such as @asteasolutions/zod-to-openapi, but creating high-quality schemas manually is still repetitive and time-consuming.

Schema Forg AI was created to automate this process using a domain-specific language model trained exclusively for API schema generation.

Training

The model was fine-tuned on thousands of synthetic examples covering:

JSON → Zod
JSON → zod-to-openapi
Zod → OpenAPI
OpenAPI → Zod
OpenAPI JSON
OpenAPI YAML

The dataset emphasizes:

Nested objects
Arrays
Enums
Validation rules
Nullable fields
Optional properties
API endpoints
Request/Response schemas
Components
Parameters

Intended Use

This model is intended for developers working with:

TypeScript
Zod
@asteasolutions/zod-to-openapi
OpenAPI
REST APIs
Backend development

Limitations

This model is highly specialized.

It performs best on API schema generation and conversion tasks and should not be considered a general-purpose coding assistant.

Author

Developed by Edson Carvalho Inturia as an educational research project exploring synthetic dataset generation, LoRA fine-tuning, and specialized language models for API development.

Downloads last month: 19

Safetensors

Model size

4B params

Tensor type

F16

Model tree for edsoncarvalhointuria/merged-schema-forg-ai

Base model

microsoft/Phi-3-mini-4k-instruct

Adapter

edsoncarvalhointuria/schema-forg-ai

Finetuned

(1)

this model

Quantizations

1 model

edsoncarvalhointuria
/

merged-schema-forg-ai