Changelog
Every public release, newest first. Each one is tested against stored plain runs before it ships.
1.0.0 today
The first public release.
Every release, newest first
- v1 (1.0.0) First public release
The first public release. One tool, model_salt, for Claude Code, Codex, Gemini CLI and Kimi Code, tested on 6 models. Ask for more or less randomness with a setting from 0 to 2. Without one it runs at 1. How it works
Right now
- Current
- v1 (1.0.0)
- Tested on
- 6 models
Claude Opus 5.5, GPT-6.1 Sol, Claude Fable 5.1, GPT-6 Astra, Gemini 4 Argon, Kimi K3
Found a prompt that still repeats?
If your model gives the same answer every time even with Model Salt, open an issue with the prompt and the plain runs. Fixes that pass the blind judge ship here with your name.
Open an issue