|
| 1 | +--- |
| 2 | +description: Roo Code 3.33.0 adds Gemini 3 Pro Preview, improves Gemini defaults and cost reporting, and enables native tool calling for the OpenAI-native provider. |
| 3 | +keywords: |
| 4 | + - roo code 3.33.0 |
| 5 | + - new features |
| 6 | + - bug fixes |
| 7 | +image: /img/v3.33.0/v3.33.0.png |
| 8 | +--- |
| 9 | + |
| 10 | +# Roo Code 3.33.0 Release Notes (2025-11-18) |
| 11 | + |
| 12 | +Gemini 3 Pro Preview is now available in Roo Code and already performing extremely well in real-world coding tasks, alongside 16 tool-protocol and UI tweaks and fixes—thanks to everyone in the Roo Code community who helped shape this release. |
| 13 | + |
| 14 | +<img src="/img/v3.33.0/v3.33.0.png" alt="Roo Code v3.33.0 Release" width="600" /> |
| 15 | + |
| 16 | +## Gemini 3 Pro Preview |
| 17 | + |
| 18 | +Roo Code now supports Google’s Gemini 3 Pro Preview model through direct Gemini, Vertex AI, and aggregator providers like OpenRouter and Requesty ([#9357](https://github.com/RooCodeInc/Roo-Code/pull/9357), [#9327](https://github.com/RooCodeInc/Roo-Code/pull/9327)): |
| 19 | + |
| 20 | +- **1M-token, reasoning-capable model**: Handles very large conversations while providing higher-quality multi-step reasoning on complex coding and refactoring tasks. |
| 21 | +- **Strong eval performance**: Achieves a 100% score on internal Roo Code evals and 76.2% on SWE-bench Verified, giving more consistent solutions on real-world coding tasks. |
| 22 | +- **Reliable tool usage**: Executes complex multi-step tool workflows without getting stuck or losing track, especially in long, tool-heavy tasks. |
| 23 | +- **Better out-of-the-box defaults**: Uses `gemini-2.5-pro` by default where supported, sets a more natural temperature of 1, cleans up the Gemini model list, and includes reasoning / “thought” tokens in cost reporting so usage numbers better match provider billing. |
| 24 | + |
| 25 | +## QOL Improvements |
| 26 | + |
| 27 | +- **Git status in environment details** ([#9310](https://github.com/RooCodeInc/Roo-Code/pull/9310)): Shows git status information in environment details so agents have more context about untracked, modified, and staged files when reasoning about your workspace. |
| 28 | +- **Tool protocol selector in advanced settings** ([#9324](https://github.com/RooCodeInc/Roo-Code/pull/9324)): Lets you choose which tool protocol to use (such as XML vs native) without editing config files, making it easier to experiment with different tool behaviors. |
| 29 | +- **Dynamic tool protocol resolution** ([#9286](https://github.com/RooCodeInc/Roo-Code/pull/9286)): Resolves the active tool protocol using a clear precedence hierarchy, so provider defaults, mode settings, and user overrides interact in a predictable way. |
| 30 | +- **Improved Modes view toolbar** ([#9077](https://github.com/RooCodeInc/Roo-Code/pull/9077)): Moves Import/Export into the Modes view toolbar and cleans up the Mode edit view, making it easier to manage and share modes from a single place. |
| 31 | +- **Cloud agent CTA points to setup page** ([#9338](https://github.com/RooCodeInc/Roo-Code/pull/9338)): Updates the cloud agent call-to-action to link directly to the setup page so new users can get started faster. |
| 32 | +- **Roo Code Cloud provider pricing page** ([#9195](https://github.com/RooCodeInc/Roo-Code/pull/9195)): Adds a pricing page and related Cloud provider tweaks so pricing is easier to understand before you enable Roo Code Cloud. |
| 33 | + |
| 34 | +## Bug Fixes |
| 35 | + |
| 36 | +- **Prevent duplicate tool_result blocks in native protocol** ([#9248](https://github.com/RooCodeInc/Roo-Code/pull/9248)): Ensures each native tool call emits a single tool_result block, avoiding 400 errors and duplicated tool executions. |
| 37 | +- **Format tool responses for native protocol** ([#9270](https://github.com/RooCodeInc/Roo-Code/pull/9270)): Normalizes the structure of tool responses so native protocol runs are easier for models to follow and less likely to error. |
| 38 | +- **Centralize toolProtocol configuration checks** ([#9279](https://github.com/RooCodeInc/Roo-Code/pull/9279)): Uses a single source of truth for toolProtocol configuration, reducing configuration drift and subtle behavior differences. |
| 39 | +- **Preserve tool blocks in conversation history** ([#9319](https://github.com/RooCodeInc/Roo-Code/pull/9319)): Keeps native protocol tool blocks intact in history so follow-up turns can reason correctly about prior tool calls. |
| 40 | +- **Prevent infinite loops after successful finalization** ([#9325](https://github.com/RooCodeInc/Roo-Code/pull/9325)): Fixes a regression where certain native tool flows could loop after successful completion instead of stopping cleanly. |
| 41 | +- **Sync parser state with profile and model changes** ([#9355](https://github.com/RooCodeInc/Roo-Code/pull/9355)): Keeps the conversation parser aligned with the active profile and model so switching models or profiles does not leave the parser in an inconsistent state. |
| 42 | +- **Pass tool protocol to truncation errors** ([#9358](https://github.com/RooCodeInc/Roo-Code/pull/9358)): Ensures truncation errors know which tool protocol is active so error handling and messaging stay accurate. |
| 43 | +- **VS Code theme-colored outline button borders** ([#9336](https://github.com/RooCodeInc/Roo-Code/pull/9336)): Aligns outline button borders with the current VS Code theme for a more consistent UI. |
| 44 | +- **Use shields.io badges instead of badgen.net** ([#9318](https://github.com/RooCodeInc/Roo-Code/pull/9318)): Replaces broken badge URLs with shields.io so badges render reliably again. |
| 45 | +- **Cap git status file sampling in evals** ([#9322](https://github.com/RooCodeInc/Roo-Code/pull/9322)): Adds a maximum for git status files in eval settings so evaluations don’t pull excessively large environment details. |
0 commit comments