AutoShortsX

Changelog

Stay up to date with the latest improvements and new features

v0.4.0 Latest
  • Fixed Groq script generation for 14-scene projects by reducing token usage to prevent rate-limit errors.
v0.2.0
  • Fixed the "Failed to validate JSON" error that blocked script generation on Groq.
  • Fixed script generation failing silently with "no content returned".
  • Removed the Llama 3.3 70B and Llama 3.1 8B models, which Groq has retired.
  • Added GPT-OSS 120B as the new default script-writing model.
  • Added GPT-OSS 20B for faster, lighter script generation.
  • Added Qwen3.6 27B as an alternate script-writing model.
  • Script generation is now dramatically faster and uses far fewer tokens per video.
  • The script writer now retries more intelligently when a model returns malformed output.
  • Error messages now explain why a model call failed instead of showing a raw API response.
  • Fixed a crash that could occur while displaying an API error message.