Models & Labs
Anthropic releases Claude Fable 5.1 with cheaper caching
Wes Rothhigh confidence
Why it matters
- →Cache read costs drop by 4x, making long-context agentic loops significantly cheaper.
- →Token efficiency improvements reduce operational expenses for high-volume automation tasks.
- →Release timing positions Anthropic competitively against OpenAI's latest frontier model announcements.
Anthropic has released Claude Fable 5.1 and Mythos 5.1, emphasizing significant improvements in long-horizon agentic work efficiency. Key updates include a fourfold reduction in prompt-cache read costs and optimized token usage for similar task outputs. The release coincides with OpenAI's announcement of its Astra model, which claims to have crossed 'Critical' cybersecurity capabilities. Independent testing demonstrates the ability to build complex applications, such as extraction games and agent simulations, within 10-15 minutes using the new effort settings.
Read originalMore in Models & Labs
Models & Labscoding
llama.cpp b10996 adds CUDA 13 and ROCm 10.0 builds
This release quietly closes the hardware gap for local inference by adding native support for CUDA 13 and ROCm 10.0 across Linux and Windows. NVIDIA users can now leverage newer driver stacks without waiting for upstream updates, while AMD GPU owners finally get first-class ROCm 10 binaries that match the maturity of their CUDA counterparts. Apple Silicon builds remain available but KleidiAI is explicitly disabled here, suggesting a focus on stability over new kernel optimizations for this specific iteration. The inclusion of openEuler support for Huawei's Ascend chips further broadens the ecosystem beyond standard x86 and ARM consumer hardware.
llama.cpp Releases
Models & Labsmodels
Llama.cpp b10981 release enhances OpenVINO support
The latest b10981 release of llama.cpp brings significant improvements to OpenVINO integration, particularly in optimizing stateful decode and GPU MoE inference. By addressing issues like stateful decode errors and enhancing the handling of sliding-window layers, this update ensures more reliable and efficient model performance. The release also introduces new features such as the GGML_OPENVINO_REQUANT_KQUANT for 4-bit requantization and improved handling of multi-head models. These changes make llama.cpp more robust and versatile, especially for developers working with complex AI models on diverse hardware setups.