Skip to content

Quick Start Guide

Get AIForge running in under 5 minutes.

1. Installation

# Core + default provider (OpenAI)
pip install aiforge-kit

# Specific providers
pip install "aiforge-kit[anthropic]"        # Anthropic + Voyage AI embeddings
pip install "aiforge-kit[local]"            # Ollama

# Everything
pip install "aiforge-kit[all]"

# Vector store backends
pip install "aiforge-kit[pinecone]"         # Pinecone
pip install "aiforge-kit[qdrant]"           # Qdrant
pip install "aiforge-kit[pgvector]"         # PostgreSQL + pgvector

# API server + Playground UI
pip install "aiforge-kit[server]"           # FastAPI + Uvicorn
pip install "aiforge-kit[playground]"       # Streamlit UI

To install the latest unreleased code instead, install from GitHub: pip install "aiforge-kit[anthropic] @ git+https://github.com/kushangpatel/aiforge.git".

Plugin System

AIForge auto-registers all 10 built-in modules and vector stores on import. You can extend it with custom plugins:

from aiforge.registry import register_module, list_modules

# Register a custom module
register_module("sentiment", SentimentAnalyzer)

# See all available modules
print(list_modules())  # ['chat', 'classify', ..., 'sentiment', 'summarize']

See the Plugin System section for full details on custom modules, providers, and vector stores.

2. Scaffold Config

aiforge init
# Creates aiforge.yaml + .env

3. Set API Key

export AIFORGE_OPENAI_API_KEY=sk-...

4. Use in Code

from aiforge import Summarizer

s = Summarizer()
result = s.run("AIForge is a plug-and-play AI toolkit for product teams...")
print(result.summary)
print(result.cost)  # Cost in USD

End-to-End Example

# 1. Install
pip install aiforge-kit

# 2. Scaffold config
aiforge init

# 3. Set API key
export AIFORGE_OPENAI_API_KEY=sk-...

# 4. Use in code
python -c "
from aiforge import Summarizer
s = Summarizer()
print(s.run('AIForge is a plug-and-play AI toolkit...').summary)
"

Module Independence

Each module is self-contained and can be used independently:

# Install only what you need
pip install aiforge-kit                        # Core + OpenAI provider
pip install "aiforge-kit[anthropic]"         # Core + Anthropic + Voyage AI
pip install "aiforge-kit[local]"               # Core + Ollama provider
pip install "aiforge-kit[all]"                 # Everything

# Use only what you need
from aiforge import Summarizer     # Just summarization
from aiforge import DocQA          # Just document Q&A
from aiforge import Pipeline       # Chain multiple modules

Try Each Module

Summarize

from aiforge import Summarizer

s = Summarizer(style="bullet_points")
result = s.run("Your long text here...")
print(result.summary)

Document Q&A (RAG)

from aiforge import DocQA

qa = DocQA(collection="my_docs")
qa.ingest("./docs/")
answer = qa.ask("How does authentication work?")
print(answer.text)
print(answer.sources)

Extract Structured Data

from aiforge import Extractor
from pydantic import BaseModel

class Invoice(BaseModel):
    vendor: str
    total: float
    currency: str

e = Extractor(output_schema=Invoice)
result = e.run("Invoice from Acme Corp. Total: $86.00 USD.")
print(result.data.vendor)  # "Acme Corp"

Classify Text

from aiforge import Classifier

c = Classifier(labels=["bug", "feature_request", "question"])
result = c.run("The login button doesn't work on mobile")
print(result.label)       # "bug"
print(result.confidence)  # 0.94

from aiforge import SemanticSearch

ss = SemanticSearch(collection="products")
ss.index([
    {"id": "1", "text": "Wireless headphones", "metadata": {"price": 299}},
    {"id": "2", "text": "Bluetooth speaker", "metadata": {"price": 79}},
])
results = ss.query("something to listen to music")
for r in results.hits:
    print(f"{r.score:.3f} | {r.text}")

Generate Content

from aiforge import Generator

g = Generator(
    template="Write a {tone} product description for {product_name}.",
    guardrails=["no_competitor_mentions"]
)
result = g.run(product_name="AIForge", tone="professional")
print(result.text)

Chat Agent

from aiforge import ChatAgent

agent = ChatAgent(system_prompt="You are a helpful assistant.")
response = agent.chat("What's the return policy?")
print(response.message)

# Multi-turn (memory maintained)
response = agent.chat("What about international orders?")

Content Moderation

from aiforge import Moderator

m = Moderator(checks=["pii", "toxicity", "prompt_injection"])
result = m.check("My email is john@example.com")
print(result.flagged)  # True

cleaned = m.redact("Call me at 555-867-5309")
print(cleaned.text)  # "Call me at [PHONE]"

Chain Modules with Pipelines

from aiforge import Pipeline, Extractor, Classifier, Summarizer

pipe = Pipeline([
    ("extract", Extractor(output_schema=TicketSchema)),
    ("classify", Classifier(labels=["critical", "high", "medium", "low"])),
    ("summarize", Summarizer(style="concise")),
])

result = pipe.run("Long customer complaint email text here...")
print(result["classify"].label)
print(result["summarize"].summary)

Launch API Server

aiforge serve --host 0.0.0.0 --port 8000

SDK or API? If your code is Python, import the SDK directly for zero overhead and full type safety. Use the HTTP API when non-Python services need access or you want a shared deployment. See the SDK vs HTTP API guide for details.

Launch Playground UI

aiforge playground
# Opens Streamlit app at http://localhost:8501

Next Steps