Search headlines, summaries and articles to connect the dots.
The qwen2vl model in llama.cpp was producing incorrect results due to NULL values being passed for FFN weight scales, preventing per-tensor NVFP4 scales from being applied. This issue has been resolved by correctly passing the scales to the build_ffn function, consistent with how qwen2 handles them.
The latest release of llama.cpp introduces support for Inkling 2, including text and vision capabilities. This update includes text inference implementation, GGUF conversion for text models, vision multimodal projection support, still-image encoding, flash attention with folded bias and sinks, and improvements to tensor mapping and testing. The changes also address model conversion, checkpoint handling, and compatibility across multiple platforms, with support for CPU, GPU (CUDA, ROCm, Vulkan), and specialized hardware like Snapdragon and OpenVINO.
The latest release of llama.cpp (b12362) includes optimizations for grammar matching through left factoring, improved handling of character class boundaries, prevention of infinite expansion, and enhanced code clarity with ASCII comments and deduplicated logic. It reverts prior changes related to AST refactoring and alternative limits, while introducing shared immutable rules and removing performance benchmarks. The release supports multiple platforms including macOS, Linux, Windows, Android, and iOS, with various backend options such as CUDA, Vulkan, ROCm, OpenVINO, and SYCL.
The update refactors DFlash-based speculative decoding methods in llama.cpp by restructuring code organization, sharing components across models, splitting decoder graphs per head, and introducing dedicated classes for DFlash2 and LiliCorr drafts. It also adds support for new draft types, improves documentation, and removes unused code.
The update requires the position embedding tensor for HunyuanVL to be explicitly provided as F32 during model loading to prevent buffer overflow issues when using F64 tensors, ensuring compatibility and stability on first image requests.
The n8n 1.123.96 update addresses a bug that made JSON highlight color invisible on light themes and includes an update to the @modelcontextprotocol/sdk.
The n8n 2.47.6 update includes bug fixes for Redis cache handling, prompt caching, legacy data reading, JSON highlighting, and Pinecone Vector Store integration. New features enable Gateway credits promotions to be displayed in the editor. Performance improvements focus on reducing runtime bundle size and optimizing expression evaluation.
The n8n 2.47.7 update resolves a bug where executions using global credentials were incorrectly redacted, ensuring proper visibility of credential data in execution logs.
The n8n 2.48.3 update addresses several bugs related to Redis caching, prompt caching, legacy data handling, and UI visibility, introduces a new feature to display Gateway credits promotions in the editor, and includes performance enhancements by optimizing character map inclusion and reducing per-item setup in the expression runtime.
The n8n 2.48.4 update resolves a bug where executions using global credentials were incorrectly redacted, ensuring proper visibility of credential data in execution logs.
The llama.cpp project has released version 1.0.0, marking a significant milestone with numerous improvements across multiple platforms including CUDA, Vulkan, Metal, SYCL, OpenCL, and WebGPU. Key updates include enhanced support for quantized models (Q5_0, Q5_1, FP8), optimizations for flash attention and MoE caching, improved server security features like CORS protection and SSRF mitigation, UI enhancements such as sidebar reorganization and live-preview markdown editing, and better cross-platform compatibility. The release also includes bug fixes, CI/CD improvements, and updated dependencies.
Release notes for n8n version 1.123.95, including a guided diff and auto-fix options via Cubic.
The n8n 2.47.5 update addresses a bug that allowed edits to read-only credential nodes via the Public API and fixes alignment issues with animated node borders in the editor.
The n8n 2.48.2 update addresses several bugs, including rejecting edits to read-only credential nodes via the Public API, aligning animated node border corners in the editor, and ensuring agent tabs remain accessible for new drafts.
The update introduces graceful failure handling for unreachable RPC endpoints and failed handshakes in ggml-rpc, replacing fatal aborts with reportable errors. Functions now return appropriate values (e.g., false, nullptr) on connection failures, allowing host applications to handle issues without crashing. A regression test was added for unreachable endpoints, and low-value comments were removed. Mid-stream failures remain unchanged to prevent silent corruption.
A bug fix in n8n version 1.123.94 ensures that retry backoff settings are preserved when a trigger fails immediately after being reactivated, improving reliability in workflow execution.
The n8n 2.47.4 update addresses several bugs, including maintaining expression transfers after built-in replacements, ensuring observational memory continues running after stopping a HITL card, matching bare node references across expression engines, optimizing font registration in the Edit Image Node, and widening the migration report notify owner column in the editor.
The n8n 2.48.1 update addresses several bugs, including issues with expression transfers after built-in replacements, matching bare node references across expression engines, optimizing font registration in the Edit Image Node, and widening the migration report notify owner column in the editor.
Article URL: https://github.com/eltmon/overdeck/releases/tag/v0.74.0 Comments URL: https://news.ycombinator.com/item?id=50450704 Points: 1 # Comments: 0
On Qualcomm Adreno GPUs, scalar flash attention was limited to 8 rows per workgroup when head size exceeds 128 to prevent register spilling and performance degradation. This change improved performance significantly, reducing execution time from several seconds to milliseconds and enabling completion of the test-backend-ops case.
