vial_llm

# Changelog

All notable changes to this project will be documented in this file.

The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).

## [0.1.0] - 2026-03-26

### Added
- Initial release as embeddable Phoenix LiveView library
- Multi-provider LLM comparison (OpenAI, Anthropic, Ollama)
- Prompt management with version control and variable interpolation
- Evaluation suites with automated test cases and assertions
- Evolution tracking to visualize prompt performance over time
- Real-time dashboard with metrics and cost tracking
- Standalone mode for running without embedding
- Mix task `mix vial.install` for easy installation
- Comprehensive test coverage and CI/CD

### Features
- **Providers**: Configure OpenAI, Anthropic, and Ollama models
- **Prompts**: Version-controlled templates with `{{variable}}` support
- **Runs**: Execute prompts across multiple providers simultaneously
- **Evaluation Suites**: Automated testing with contains/regex/exact_match assertions
- **Evolution View**: Track prompt performance across versions and providers
- **Dashboard**: Live metrics, cost trends, and latency tracking
- **Embeddable**: Mount in any Phoenix LiveView app as a self-contained dashboard
- **Customizable**: Resolver pattern for access control and user identification

[0.1.0]: https://github.com/ccarvalho-eng/vial/releases/tag/v0.1.0