Cline v4.0.2 Adds DeepSeek Reasoning Effort Controls, Including 'xhigh' Setting
Cline, the autonomous AI coding assistant, released v4.0.2 with support for reasoning effort controls on DeepSeek thinking models, including the new 'xhigh' setting. The update also improves the ClinePass provider experience with clearer reasoning controls and model selection.
Cline v4.0.2 Adds DeepSeek Reasoning Effort Controls
Cline, the autonomous coding assistant extension, released version 4.0.2 on June 29, 2024, adding support for reasoning effort controls on DeepSeek thinking models.
Key Updates
The release adds reasoning effort support for DeepSeek's thinking models, including the new "xhigh" setting. This allows users to control the computational effort DeepSeek's reasoning models apply to tasks, with higher effort levels producing more thorough chain-of-thought responses.
The update improves the ClinePass provider experience with clearer reasoning controls and model selection. ClinePass is Cline's proprietary API service that provides access to various AI models.
Technical Fixes
The release includes several technical improvements:
- Reasoning effort controls now display correctly for ClinePass models
- ClinePass model resolution has been aligned with other providers
- The extension now uses canonical Cline Z.ai model IDs
- Environment variable replacement in the webview has been fixed
- Focus chain settings now default correctly in webview state, ensuring the toggle reflects accurate values on load
ClinePass and Z.ai model metadata has been polished for better clarity.
What This Means
This release positions Cline to take advantage of DeepSeek's reasoning capabilities, particularly for complex coding tasks that benefit from extended chain-of-thought processing. The "xhigh" effort setting suggests DeepSeek has expanded its reasoning effort options beyond the standard tiers, potentially offering more granular control over compute-intensive reasoning tasks. The focus on improving provider experience and model metadata indicates Cline is standardizing its multi-model support as the number of available AI coding models continues to expand.
Related Articles
AWS Publishes Reference Architecture for Multimodal WhatsApp Ordering Agents Using Bedrock AgentCore and Nova 2
AWS published a reference architecture showing how to deploy a WhatsApp ordering assistant on Amazon Bedrock AgentCore, using Nova 2 Lite for text and Nova 2 Sonic for voice, with shared cross-channel memory and MCP-based tool access to backend systems.
Gemini Overlay on Android Adds Minimize Button for Multitasking Bubble
Google is widely rolling out a new Minimize button for the Gemini overlay on Android, which collapses conversations into a floating bubble users can drag or tap to expand. The feature currently supports six fixed positions and has a reported bug that resets bubble placement after each minimization.
OpenAI Lists GPT-6 Astra Pro on OpenRouter: Same Model, Higher-Compute Reasoning Mode
GPT-6 Astra Pro, now listed on OpenRouter, is the existing GPT-6 Astra model configured to run with reasoning.mode set to 'pro' for higher-quality output on complex tasks. It carries a 1M-token context window and tiered pricing from $5/$25 to $20/$100 per million input/output tokens depending on the serving tier.
xAI Brings Grok Bot to iPad and Android, Cuts Price From $300/Month to $20/Month
xAI has expanded its Grok Bot AI agent app from iPhone and Mac to iPad and Android, while slashing the price from $300/month to compatibility with the $20/month Cursor Pro plan. Enterprise customers using Grok or Cursor get free access for a limited time.
Comments
Loading...