WHITTL
A Python development environment with built-in AI assistance, one-click builds for Windows and Linux (plus experimental Android APK support), and a full-featured code editor. Write code, generate code, or both.
IMAGE → APP
Drop a screenshot of any UI into Whittl and get a native Python app. A Figma mock, a web app you want to run locally, your own music player you're tired of configuring. Every other AI coding tool outputs more web. Whittl uniquely converts design references into runnable software you own.
- Drag-drop or clipboard-paste a screenshot into chat
- Works with Claude (Opus / Sonnet / Haiku), GPT-4o, Gemini 2.5 Pro/Flash, Llama 3.2 Vision, Pixtral, Qwen-VL, Gemma 3 — any vision-capable model on OpenRouter
- Images flow through tool-use rounds, not just the initial chat turn — the AI sees your image while it builds, not just when you send it
- Preview before send — attached images show inline with filename and size, removable, no silent bytes leaving your machine
DEVELOPMENT TOOLS
- Code editor with syntax highlighting, AST-powered code folding, autocomplete, and bracket auto-pairing
- Real-time error detection (pyflakes) with inline error indicators
- Multi-file project management with tabbed editing and line count tracking
- Smart file routing — keyword search + AI selects the right files in large projects
- One-click builds — Windows .exe, Linux AppImage, Android APK, and setup installers
- Explain mode — highlight any code and get a plain-English explanation
- Undo/history — roll back bad generations to any previous state
- Multi-file search across your entire project
- Auto-update check — get notified when a new version is available
- Real .py files on your disk — open in any editor, version with git, own forever
- Works on Windows and Linux
AI ASSISTANCE
- 4 targets — Desktop (PySide6), Mobile (Flet), Basic (CustomTkinter), or General Python (CLI tools, scripts, plugins)
- Conversational AI — ask questions, get explanations, iterate on your code through chat
- 5 AI backends — Claude, DeepSeek, Gemini, OpenRouter (200+ models), or fully offline with Ollama
- Smart editing — the AI changes only the lines that need to change, dramatically faster and cheaper than rewriting whole files
- Self-healing autofix — when your app crashes, the AI reads the traceback and iterates on fixes automatically, up to 10 rounds until it works
- Built-in cost cache — multi-round conversations reuse context so your API bill stays small
- 75+ auto-fix rules catch common mistakes — hallucinated APIs, deprecated patterns, missing imports
- Context memory — the AI remembers your earlier conversation instead of forgetting after a few messages
- Bring your own API key — no middleman markup, no message limits, no monthly quotas
MOBILE BUILDS EXPERIMENTAL
Mobile APK builds are currently experimental. They work but may need troubleshooting on some devices. Desktop builds (Windows/Linux) are stable and production-ready.
- Build Android APKs with one click — no Android Studio, no Java, no Gradle
- Device format selector — Phone, Tablet, or custom dimensions
- One-click install to device via USB
- Pure Python APK signing — all build tools downloaded automatically
SCREENSHOTS
MOBILE APP OUTPUT
Phone — Recipe App
Tablet — Same App
DARK & LIGHT THEMES
v2.3's brand pass brought Cormorant Garamond and Inter typography, Phosphor Duotone icons throughout, and a refined navy, cream, tan, and copper palette. Dark and light themes are both polished, and flip cleanly without stale colors bleeding through.
DEMO
HOW IT WORKS
1. Start a new project or import existing Python code
2. Write code in the editor, or describe what you want and let AI generate it
3. Use the AI chat to ask questions, explain code, or iterate on changes
4. Run and test directly inside Whittl with built-in error detection
5. Build to .exe, AppImage, or Android APK with one click
What's New in v2.3.0
- Screenshot to App. Drop a screenshot of any UI into chat and a vision-capable model rebuilds it as a native Python app. Works across Claude, GPT-4o, Gemini 2.5, Llama 3.2 Vision, Pixtral, Qwen-VL, Gemma 3, and every vision-capable model on OpenRouter's catalog. Full multimodal flow works through tool-use rounds, not just the first message.
- Agent Mode. Opt-in unbounded tool-use loop for capable models (Claude Opus/Sonnet, GPT-4o, Gemini 2.5 Pro, Qwen3.5-Plus, DeepSeek V3.2) with a real shell tool, cross-turn session memory, and hard safeguards so weak models can't burn tokens in loops they can't solve.
- The Brand Pass. Cormorant Garamond and Inter typography, Phosphor Duotone icons replacing emoji, refined navy/cream/tan/copper palette. Chat has real avatars. Dark and light themes both polished and flip cleanly without stale colors.
- Auto-fix Safeguards. Oscillation guard detects when the AI is stuck bouncing between two errors. Hard 5-round cap stops runaway loops cost dead. Stop button now actually cancels queued rounds instead of flipping to Generate between them.
- Capability Chips. Models dialog shows concrete
[Tools] [Thinks] [Long] [Vision]chips per model instead of opaque S/A/B tier letters. You can verify what a model actually supports before picking it. - Multi-file Live Streaming. Multi-file generations open each file in its own tab as the AI writes it. Code streams in character-by-character, not 5-line jumps. No more waiting for everything to finish before seeing results.
- Expanded Project Icons. 125+ Phosphor Duotone SVGs covering weather, cameras, mountains, books, charts, terminals, and more. Keyword matcher auto-picks an icon per project from its name; right-click to override.
- Paid-user Downgrade Fix. A startup migration used to silently rewrite paid OpenRouter model IDs to their
:freevariants on every launch. Paid users who picked a paid model were getting routed through the rate-limited free tier without knowing. Removed. - OpenRouter Favorites That Persist. Starred models now actually stick between sessions. Fixed a race between the background update-checker and the favorites writer.
- 938 Unit Tests. Up from 683 in v2.2 (+255 tests). Heavy focus on safeguards that matter: oscillation detection, round cap, theme-flip regressions, streaming parser edge cases, favorites race, project icon availability.
FAQ
Is Whittl a one-time purchase?
Yes. $20 CAD one-time with no subscription, no message limits, and no monthly fees.
What AI providers does Whittl support?
Claude, Gemini, DeepSeek, OpenRouter (200+ models including free ones), and Ollama for fully offline use. Bring your own API key.
Can Whittl build Android apps?
Yes, though Android builds are currently experimental. One-click APK builds with no Android Studio or Java required. Desktop builds are stable and production-ready.
Do I own the code Whittl generates?
Yes. All code is saved as real .py files on disk. Open in any editor, version with git, own forever.
YOU MIGHT ALSO LIKE
Hybrid pixel and vector animation tool for game assets and illustration.
Fast, lightweight RAW image editor for photographers.