Skip to main content

Roo Code 3.29.4 Release Notes (2025-10-31)

This release adds the MiniMax provider, improves @ file search on large projects, clarifies the Logs tab and VS Code LM guidance, and delivers multiple stability fixes.

Roo Code v3.29.4 Release

We’re hiring​

We added a “We’re hiring” link to the announcement modal. Explore open roles at careers.roocode.com.

MiniMax provider​

We've added the MiniMax provider so you can run coding and reasoning tasks with strong Chinese language support and competitive pricing (thanks Maosghoul!) (#8820):

QOL Improvements​

  • Improve @ file search for large projects with a higher default index limit and respect for VS Code ignore settings; add the roo-cline.maximumIndexedFilesForFileSearch setting for tunable limits (thanks Naituw!) (#8805)
  • Rename MCP “Errors” tab to “Logs” to match mixed-level messages; clearer empty state (“No logs yet”) (#8894)
  • Custom modes now load from your configured storage path and persist after restart (thanks elianiva!) (#8499)
  • Breaking: Removed search_and_replace tool. Use apply_diff or insert_content instead. Migration notes below. (#8892)
  • Clarified VS Code LM API integration warning in Settings to reduce “model not supported” errors (#8493)

Bug Fixes​

  • Reasoning effort selection now auto-enables reasoning when needed so UI and behavior stay in sync (#8890)
  • Reduce noisy cloud-agent exceptions by suppressing repeated auth messages (#8577)
  • Prevent MCP server restart when toggling “Always allow” for MCP tools (thanks heyseth!) (#8633)
  • Reuse existing Qdrant index after outages to avoid full reindex and cut restart time (thanks heyseth, jaroslaw-weber!) (#8588)
  • Make code index initialization non‑blocking at activation to avoid startup hangs (thanks cjlawson02!) (#8933)
  • Honor maxReadFileLine across code definition listing and file reads to prevent context overflows (thanks elianiva, chenxluo!) (#8509)
  • Prevent infinite retry loop when canceling during auto‑retry (thanks mini2s!) (#8902)
  • Gate auth‑driven Roo model refresh to the active provider only to reduce background work (#8915)

Provider Updates​

  • Cerebras: add zai‑glm‑4.6 and change default to gpt‑oss‑120b; deprecate qwen‑3‑coder models (#8920)