* feat: add zai-glm-4.6 model and update gpt-oss-120b for Cerebras
- Add zai-glm-4.6 with 128K context window and 40K max tokens
- Set zai-glm-4.6 as default Cerebras model
- Update gpt-oss-120b to 128K context and 40K max tokens
* feat: add zai-glm-4.6 model to Cerebras provider
- Add zai-glm-4.6 with 128K context window and 40K max tokens
- Set zai-glm-4.6 as default Cerebras model
- Model provides ~2000 tokens/s for general-purpose tasks
* add [SOON TO BE DEPRECATED] warning for Q3C
* chore: set gpt-oss-120b as default Cerebras model
* Fix cerebras test: update expected default model to gpt-oss-120b
* Apply suggestion from @mrubens
Co-authored-by: Matt Rubens <mrubens@users.noreply.github.com>
---------
Co-authored-by: kevint-cerebras <kevin.taylor@cerebras.net>
Co-authored-by: Matt Rubens <mrubens@users.noreply.github.com>
* fix: prevent MCP server restart when toggling tool permissions
Add isProgrammaticUpdate flag to distinguish between programmatic config
updates and user-initiated file changes. Skip file watcher processing
during programmatic updates to prevent unnecessary server restarts.
* fix(mcp): prevent server reconnection when toggling disabled state
Fixed bug where MCP servers would reconnect instead of staying disabled when toggled off. The issue was that toggleServerDisabled() used stale in-memory config instead of reading the fresh config from disk after writing the disabled flag.
Changes:
Added readServerConfigFromFile() helper to read and validate server config from disk
Updated disable path to read fresh config before calling connectToServer()
Updated enable path to read fresh config before calling connectToServer()
This ensures the disabled: true flag is properly read, causing connectToServer() to create a disabled placeholder connection instead of actually connecting the server.
+ refactor(mcp): use safeWriteJson for atomic config writes
Replace JSON.stringify + fs.writeFile with safeWriteJson in McpHub.ts
to prevent data corruption through atomic writes with file locking.
* fix(mcp): prevent race condition in isProgrammaticUpdate flag
Replace multiple independent reset timers with a single timer that gets
cleared and rescheduled on each programmatic config update. This prevents
the flag from being reset prematurely when multiple rapid updates occur,
which could cause unwanted server restarts during the file watcher's
debounce period.
+ fix(mcp): ensure isProgrammaticUpdate flag cleanup with try-finally
Wrap safeWriteJson() calls in try-finally blocks to guarantee the
isProgrammaticUpdate flag is always reset, even if the write operation
fails. This prevents the flag from being stuck at true indefinitely,
which would cause subsequent user-initiated config changes to be
silently ignored.
* feat: Add supportsReasoning property for Z.ai GLM binary thinking mode
- Add supportsReasoning to ModelInfo schema for binary reasoning models
- Update GLM-4.5 and GLM-4.6 models to use supportsReasoning: true
- Implement thinking parameter support in ZAiHandler for Deep Thinking API
- Update ThinkingBudget component to show simple toggle for supportsReasoning models
- Add comprehensive tests for binary reasoning functionality
Closes#8465
* refactor: rename supportsReasoning to supportsReasoningBinary for clarity
- Rename supportsReasoning -> supportsReasoningBinary in model schema
- Update Z.AI GLM model configurations to use supportsReasoningBinary
- Update Z.AI provider logic in createStream and completePrompt methods
- Update ThinkingBudget UI component and tests
- Update all test comments and expectations
This change improves naming clarity by distinguishing between:
- supportsReasoningBinary: Simple on/off reasoning toggle
- supportsReasoningBudget: Advanced reasoning with token budget controls
- supportsReasoningEffort: Advanced reasoning with effort levels
* Use monotonic clock for rate limiting
* Fix LiteLLM test failures after merge
- Remove supportsComputerUse from LiteLLM implementation as it's no longer part of ModelInfo interface
- Update test expectations to include cacheWritesPrice and cacheReadsPrice fields
- Fix test for max_output_tokens preference functionality
---------
Co-authored-by: Christiaan Arnoldus <christiaan.arnoldus@outlook.com>
* changeset version bump
* Update CHANGELOG for version 3.29.1 release
Updated version number and added release notes for 3.29.1.
---------
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: Matt Rubens <mrubens@users.noreply.github.com>
* feat(zai): add GLM-4.5-X, AirX, Flash; sync with Z.ai docs; keep canonical api line keys
* feat(zai): add GLM-4.5V vision model (supportsImages, pricing, 16K max output); add tests
* feat(types,zai): sync Z.AI international model map and tests
- Update pricing, context window, and capabilities for:
glm-4.5-x, glm-4.5-airx, glm-4.5-flash, glm-4.5v, glm-4.6
- Add glm-4-32b-0414-128k
- Align tests with new model specs
* fix(zai): align handler generics with expanded model ids to satisfy CI compile step
* chore(zai): remove tier pricing blocks for Z.ai models
* fix(zai): simplify names in zaiApiLineConfigs for clarity
* chore(zai): set default temperature to 0.6
---------
Co-authored-by: Roo Code <roomote@roocode.com>
chore(gpt5): stop persisting instructions/reasoning_summary in UI message metadata
Problem: ui_messages.json was getting bloated with unused or duplicated content (system 'instructions' and 'reasoning_summary') that we do not read back. Root cause: earlier OpenAI Responses API implementation persisted these fields to per-message metadata; however, 'instructions' are already sent as top-level request instructions and 'reasoning_summary' is surfaced live via streaming events. Neither field is consumed from storage. Changes: (1) Task.persistGpt5Metadata now stores only previous_response_id; (2) removed instructions and reasoning_summary from types; (3) updated Zod schema; (4) persistence layer writes messages as-is (no sanitizer); (5) tests green. Impact: smaller ui_messages.json, no runtime behavior change for requests. Migration: old metadata fields will be ignored by schema.
* fix(editor): prevent file editing issues when git diff views are open
Add scheme checks to ensure only file:// URIs are matched when finding editors,
avoiding issues with git diffs and other schemes. Includes error logging for
failed editor lookups.
* Remove the warnings
* fix(editor): enforce file:// scheme in editor lookups to prevent git diff issues
---------
Co-authored-by: daniel-lxs <ricciodaniel98@gmail.com>
