Anthropic has released Claude Code v2.1.282, a patch focused on stability and session integrity rather than new features. Key fixes include resolving issues where resumed conversations corrupted API requests by re-sending messages, and addressing login refresh loops that blocked multi-window workflows. The update also corrects managed setting parsing errors in enterprise environments and improves error handling for third-party gateway integrations. Additional changes refine the terminal UI, fixing vim mode behaviors and rendering glitches in fullscreen modes.
Read originalThis release quietly solves a major pain point for enterprise AI workflows by adding gateway hint headers, allowing LLM gateways to correctly group requests per user prompt instead of treating them as isolated events. The new managed settings for availableModelsMatch and deniedModels give organizations precise control over model access, blocking specific versions even when broader allowances exist. Beyond governance, the update stabilizes the plugin ecosystem with rigorous validation checks that prevent silent failures from broken or misconfigured extensions. These changes shift Claude Code from a developer tool to a manageable enterprise component.
This release stabilizes Claude Code's core reliability by fixing persistent bugs in session resumption and prompt caching that previously caused data loss or infinite loops. It also tightens enterprise security with new Bedrock upstream support for IAM role assumption and mandatory guardrail application. The update addresses critical edge cases like proxy stream drops and oversized tool calls, ensuring smoother operation in complex development environments.
This release quietly refactors how llama.cpp handles Flash Attention on Apple Silicon by splitting kernels into per-dtype libraries. It’s a structural optimization that likely reduces memory overhead and improves compilation times for Metal users, though the immediate performance gains are subtle compared to algorithmic leaps. The build matrix remains massive, adding ROCm 10.0 and CUDA 13.4 support while disabling KleidiAI on Apple Silicon for now. This is infrastructure maintenance rather than a feature breakthrough, but it keeps the runtime robust across the expanding landscape of hardware backends.
This release tackles a specific performance bottleneck on Apple Silicon by extending Metal FWHT kernels to handle block widths up to 8192. Previously limited to 512, these wider operations now use threadgroup memory instead of registers, enabling faster inference for larger context windows or model architectures that rely on Hadamard transforms. The update also ships binaries for CUDA 13 and ROCm 10.0, keeping the toolkit aligned with the latest NVIDIA and AMD driver ecosystems. It’s a quiet but necessary optimization that improves throughput on M-series chips without changing the user experience.
One of llama.cpp's quietest but biggest releases. With ROCm 10.0 added as a default build, AMD GPU users stop being second-class citizens for local inference — the gap with CUDA narrows visibly. Apple Silicon Macs now compile in KleidiAI by default, meaning every M-series machine gets ARM-tuned GEMM kernels for free, no flag-flipping required. There's no new model and no new quantization here — just llama.cpp quietly becoming the inference runtime for everyone who isn't on NVIDIA. That's the headline.