README
¶
GoChat Examples
This directory contains practical examples demonstrating how to use the GoChat library.
Running the Examples
All examples require API keys to be set as environment variables:
export OPENAI_API_KEY="your-key-here"
export ANTHROPIC_API_KEY="your-key-here" # For Anthropic examples
Then run any example:
go run examples/01_basic_chat/main.go
Examples
01_basic_chat
The simplest possible usage: send a message, get a response.
What you'll learn:
- Creating a client
- Sending a single message
- Getting a response
- Accessing token usage
02_multi_turn
Maintaining conversation history across multiple turns.
What you'll learn:
- Building conversation history
- Using system messages
- Context retention across turns
- Appending assistant responses to history
03_streaming
Getting responses token-by-token as they're generated.
What you'll learn:
- Using
ChatStreaminstead ofChat - Iterating through stream events
- Handling streaming errors
- Getting usage info after streaming completes
04_tool_calling
Allowing the model to call external functions.
What you'll learn:
- Defining tools with JSON Schema
- Detecting when the model wants to call a tool
- Executing tools and returning results
- Multi-turn tool calling flow
05_multiple_providers
Using different LLM providers with the same code.
What you'll learn:
- Provider-agnostic code using
core.Clientinterface - Switching between OpenAI, Anthropic, and Ollama
- Provider-specific configuration
06_image_input
Sending images to vision-capable models.
What you'll learn:
- Reading and encoding images to base64
- Creating multimodal messages (text + image)
- Using vision models (GPT-4 Vision, Claude 3)
- Handling different image formats
07_document_analysis
Analyzing text documents (code, markdown, text files).
What you'll learn:
- Reading file content
- Sending documents for analysis
- Structured analysis prompts
- Token usage for large inputs
08_multiple_images
Analyzing multiple images in one request.
What you'll learn:
- Sending multiple images in one message
- Batch image processing
- Comparing and contrasting images
- Building complex multimodal messages
09_helper_utilities
Reusable helper functions for common tasks.
What you'll learn:
- Loading images as ContentBlocks
- Creating multimodal messages easily
- Managing conversation history
- Adding tool results to conversations
- Best practices for code reuse
More Examples
For more advanced examples, see the main README.md:
- Multimodal inputs (images)
- Custom retry strategies
- Error handling patterns
- Token usage monitoring
- Extended thinking (o1/o3 models)
Directories
¶
| Path | Synopsis |
|---|---|
|
Basic example demonstrating simple chat completion with OpenAI.
|
Basic example demonstrating simple chat completion with OpenAI. |
|
Multi-turn conversation example.
|
Multi-turn conversation example. |
|
Streaming response example.
|
Streaming response example. |
|
Tool calling example.
|
Tool calling example. |
|
Multiple providers example.
|
Multiple providers example. |
|
Multimodal input example - sending images to the model.
|
Multimodal input example - sending images to the model. |
|
Document analysis example - analyzing PDF, text files, or other documents.
|
Document analysis example - analyzing PDF, text files, or other documents. |
|
Multiple images example - analyzing multiple images in one request.
|
Multiple images example - analyzing multiple images in one request. |
|
Helper utilities for common use cases.
|
Helper utilities for common use cases. |