Intro
The AI tooling landscape has exploded recently, and it can be overwhelming to sift through the noise. This review covers some tools I've found genuinely handy. There's also a really nice PDF report I've attached below, worth a read if you have the time.
AI Tooling Review and Analysis
Search Engines
perplexity.ai
Perplexity.ai excels at delivering concise answers with source citations, making it ideal for research and fact-checking. It blew my mind when I first started using it a few month back. It's magic yo.
Gemini
Google's Gemini offers a more integrated search experience, blending traditional search results with AI-powered summaries. Its deep integration with other Google services is a significant advantage (if you're ok with that...). I'd stick with the Pro model, as Flash tends to be good for simpler tasks. The live mode is downright creepy with how real it sounds and responds.
phind.com
Phind works as a great alternative and second source to engines like Gemini and Perplexity. I do wish I used it more, but it's usually worth a shot if the other two aren't cutting it.
AI Tools
I've also dug through a fair bit a tooling as I often do. I've had, of all things, Gemini fill in the gaps below, as I've been pretty busy with work. Aichat in particular, I'd say check out out. It's better then Goose, but the LTM stuff with Pieces is nuts with how it sort of follows you around, taking notes on your work.
LTM (Long-Term Memory) - Pieces
Pieces is a powerful CLI and GUI tool for saving and managing code snippets. Its AI capabilities allow for semantic search and intelligent tagging, making it easy to find and reuse code.
Links & Stats
👉 https://github.com/pieces-app/cli-agent
AI Agent - Goose
Goose is a versatile AI agent designed for automation and task execution. While still under development, its potential for streamlining workflows and automating repetitive tasks is significant.
Links & Stats
👉 https://github.com/block/goose
AI Chat - aichat
aichat provides a convenient CLI for interacting with various large language models. Its simple interface makes it easy to experiment with different models and access their capabilities from the terminal.
Links & Stats
👉 https://github.com/sigoden/aichat
GUI - Jan
Jan offers a clean and intuitive GUI for interacting with LLMs. Its cross-platform compatibility and user-friendly design make it accessible to a wide range of users.
Links & Stats
👉 https://github.com/menloresearch/jan
IDE Plugin - ProxyAI
ProxyAI integrates AI assistance directly into your IDE, providing code completion, generation, and refactoring suggestions. This tight integration can significantly speed up development workflows.
Links & Stats
👉 https://github.com/carlrobertoh/ProxyAI
Model - Qwen2.5-Coder
Qwen2.5-Coder is a specialized language model designed for code generation and understanding. Its performance on coding tasks makes it a valuable tool for developers.
Links & Stats
👉 https://github.com/QwenLM/Qwen2.5-Coder
Honorable Mentions (Local Based)
Running AI tools locally offers benefits in terms of privacy, security, and customization. These options are worth exploring:
| Platform | App | Link |
|---|---|---|
| IDE (JetBrains + VSCode) | Tabby | https://www.tabbyml.com |
| IDE (VSCode) | Privy | https://marketplace.visualstudio.com/items?itemName=Privy.privy-vscode |
| IDE (VSCode) | Ollama Autocoder | https://marketplace.visualstudio.com/items?itemName=10nates.ollama-autocoder |
| Server | Ollama | https://ollama.com/ |
| Server | llama.cpp | https://github.com/ggerganov/llama.cpp |
Conclusion
The AI tools discussed here represent a powerful and rapidly evolving landscape. By integrating these tools into your workflow, you can significantly enhance your productivity and unlock new possibilities. This is just the beginning, and I'm excited to see what the future holds for AI-powered tooling.