AI Briefing#bug-fix
The latest AI and developer news about #bug-fix, with the original source and a short summary.
The latest AI and developer news about #bug-fix, with the original source and a short summary.
The built-in mod, enabled by default in Claude Code v2.1.317 and later, deterministically blocks the model from including user email addresses in web requests.
The patch release includes 38 pull requests that fix Git rollback failures and stale search index repairs.
The patch release from 163 contributors resolves issues preventing installation, updates, and reply delivery without adding new features.
The patch release resolves 61 pull requests to fix Gateway update checks, preserve agent conversation context, and correct macOS Talk and Bun startup bugs.
The patch resolves false offline maintenance errors and Windows setup issues that blocked fresh installations and Doctor runs.
The fix resolved a rare empty-response bug caused by capability mismatches in shared CDN caches, achieving a 99.99% success rate.
A community-developed Jinja template for GPT-OSS fixes a bug in the widely used Unsloth version that caused the model to drop final answers when reasoning traces were present in chat history.
Released the open-source tool entail to verify and prevent the issue where vLLM's rope_scaling override bug causes GSM8K scores to drop by up to 28%.
The latest Claude Code update introduces memory usage warnings, a new environment variable for controlling MCP server connection wait times, and various bug fixes and stability improvements.
vLLM resolved the non-deterministic output issue for Qwen3.8-Flash-Next at temperature 0 by fixing the sparse-attention kernel.
Anthropic resolved a bug that was abnormally depleting the weekly usage limits of Claude Code Pro and Max users and reset the limits.
Unsloth fixed the Mistral Medium 3.5 inference bug and updated the GGUF.
A parser error in vLLM was found to cause malfunctioning Tool Calling functionality in Qwen models.
A bug that caused Opus 4.7's 1M context to be truncated to 200K has been fixed.
If the gateway won't come up after the 4.14 update, suspect a tangled entrypoint path.
E16's middle-ellipsis text-fitting logic fell into an infinite loop due to Newton-style estimation getting stuck oscillating.