Tag
Claude Devs reset 5-hour and weekly rate limits for Pro and Max users, and fixed an issue causing excessive parallel subagents in Claude Code sessions.
Peter Steinberger used Codex to build a fully automated QA bot that automatically generates tests, runs tests after each code commit, and can automatically fix bugs and submit PRs, greatly improving development efficiency.
Llama.cpp release B9406 fixes a crash (GGML_ASSERT) when using MTP with MoE vision models like Qwen3.6-35B-A3B.
llm_wiki v0.4.13 released, primarily fixes bugs and merges community PRs, project stars exceed 9k+.
A patch for Firefox to work around a crash on Intel Raptor Lake CPUs, authored by glandium.
A developer uses LLMs and algebraic reformulation to formally verify a bug fix for the 2023 UK air traffic control meltdown in the Lean proof assistant, finding that LLMs are great at grinding proofs but poor at specifications.
A bug in the Windows debugging engine for x86-64 incorrectly reported the parity flag, going unnoticed for over two decades, highlighting how rarely the flag is debugged.
slock has recently updated many practical features and fixed several bugs. The author recommends users update to the latest version to experience it.
GPT-5.5 experienced capability degradation in Codex over approximately 48 hours. Two issues were found and fixed; usage limits will be reset after monitoring confirms full recovery.
llama.cpp b9158 has been released with a fix for Flash Attention on RDNA3 GPUs, improving performance for AMD users.
Qwen3.6-27B-FP8 model is now running on Intel Arc Pro B70 GPUs at ~50 tok/s with a vLLM bug fix, marking a significant milestone for Intel GPU local AI inference.
Cloudflare details a bug in their QUIC implementation, quiche, where a Linux kernel optimization for CUBIC congestion control caused performance issues, and describes the resulting fix.
Upcoming updates address crashes between multimodal projection and multi-token prediction by enabling image processing through draft contexts. The changes also introduce parallel draft support to improve speculative decoding scalability.
A corrected 2-bit GGUF model file has been uploaded to Hugging Face after fixing a bug in the imatrix computation, leading to improved logits recall and reduced error.
A game developer describes fixing a GPU rendering bug in their game Blackshift, where float precision issues when casting 8-bit adjacency integers to floats caused visual artifacts on certain NVIDIA GPUs, with the bug appearing in the main render but not in preview mode.
The article details a technical investigation into how an HTTP header caused time skew on the time.gov website, explaining the differences between NTP and HTTP time synchronization.
This repository provides fixed Jinja chat templates for Qwen 3.5 and 3.6, addressing rendering errors, token waste, and missing features in the official templates for engines like LM Studio and llama.cpp.
Apple patched an iOS bug that let law enforcement extract deleted Signal messages via cached notifications retained up to a month, after 404 Media revealed the FBI had exploited it.
Anthropic released a postmortem detailing three infrastructure bugs between August and September that intermittently degraded Claude's response quality. The report explains the technical causes, including context window routing errors, and outlines measures to prevent future incidents.
Anthropic released a postmortem addressing recent quality reports for Claude Code, identifying and fixing three issues related to reasoning effort defaults, session state management, and system prompts that affected Sonnet and Opus models.