Cline v3.89.1 Fixes Anthropic Provider Breakage on VS Code 1.123+, Adds DeepSeek V4 Support
Cline released version 3.89.1 on June 11, addressing a critical bug where the Anthropic provider broke on VS Code 1.123 and later due to Node 24 runtime incompatibility. The update also adds support for DeepSeek V4's reasoning format.
Cline v3.89.1 Fixes Anthropic Provider Breakage on VS Code 1.123+, Adds DeepSeek V4 Support
Cline, the AI coding assistant extension, released version 3.89.1 on June 11, 2025, addressing a critical compatibility issue with Anthropic's API provider and adding support for DeepSeek's latest reasoning model.
Key Fixes
The primary fix restores functionality for the Anthropic provider on VS Code version 1.123 and later. According to the release notes, VS Code's update to the Node 24 runtime broke the bundled Anthropic SDK that Cline relies on. This would have prevented users on the latest VS Code releases from accessing Claude models through Cline.
The update also adds handling for DeepSeek V4's reasoning format, expanding compatibility with the Chinese AI lab's latest model that features extended chain-of-thought capabilities.
Technical Context
The Node 24 runtime update in VS Code 1.123 introduced breaking changes that affected how bundled SDKs operate within VS Code extensions. This type of runtime upgrade can cause dependency conflicts when extensions package specific SDK versions that rely on older Node.js features or APIs.
The DeepSeek V4 addition suggests Cline is actively maintaining compatibility with newer reasoning models that output structured thinking processes before final answers, similar to OpenAI's o1 series.
What This Means
This is a maintenance release addressing ecosystem compatibility rather than adding new features. For developers using Cline with Claude models on recent VS Code versions, this patch is essential to restore functionality. The quick turnaround on the Node 24 compatibility issue demonstrates responsive maintenance, though it also highlights the fragility of extension ecosystems when core runtime dependencies change.
The DeepSeek V4 support indicates Cline is tracking new reasoning model formats as they emerge, which will be important as more providers adopt extended thinking modes in their models. Users should update to avoid service disruptions if running VS Code 1.123 or later.
Related Articles
Cline v4.0.11 Adds Claude Opus 5 and Moonshot Kimi K3 Support, Fixes Overstated 1M-Context Pricing
Cline's v4.0.11 release adds provider support for Claude Opus 5 (including 1M context window variants) across six integrations and introduces Moonshot Kimi K3 support. The update also fixes a pricing bug that overstated costs for Opus 1M-context requests above 200k tokens.
Claude Opus 5 Scores 30.2% on ARC-AGI-3, Nearly 4x the Previous Record
Claude Opus 5 scored 30.2 percent on the ARC-AGI-3 benchmark, nearly four times the previous record of 7.8 percent set by OpenAI's GPT-5.6 Sol (Max). The ARC Prize team attributes the leap to genuinely stronger reasoning, though an independent test on a separate puzzle benchmark showed far smaller improvements.
Anthropic's Claude Opus 5 Hits 0% Prompt Injection Success Rate in Browser Agent Tests, With Defenses Enabled
Anthropic's system card for Claude Opus 5 reports a 0% prompt injection success rate across 129 browser agent test scenarios when Auto Mode is enabled. On Gray Swan's broader indirect prompt injection benchmark, Opus 5 posted a 2.0% attacker success rate after 15 attempts, the lowest among tested frontier models.
Claude Opus 5 Scores 61 on Intelligence Index, Beats Fable 5 on Cost Across Most Benchmarks
Anthropic's Claude Opus 5 posts a 61 on the Artificial Analysis Intelligence Index, narrowly beating Claude Fable 5 (60) and GPT-5.6 Sol (59) while costing less per task. The model leads in coding and knowledge-work benchmarks but shows a rising hallucination rate of 50 percent.
Comments
Loading...