memory
Found in 4 packages: llama-cpp, vercel-functions, vercel-cli, crewai
llama-cpp(2 releases)
b10690This release addresses a memory-related crash during context shifts for unquantized K caches. It ensures the Hadamard matrix is copied to the k_rot tensor only when a buffer is assigned, preventing potential issues.
b7422BreakingThis release introduces graph reuse for SSM, hybrid, and recurrent models to optimize memory and performance. It also announces a forthcoming change in Linux distribution formats from .zip to .tar.gz.
vercel-functions(2 releases)
vercel@54.18.1This patch fixes an issue where per-service function configurations were ignored during the build process for experimental V2 services. It also scopes derived queue consumer groups by service name to prevent path collisions.
@vercel/build-utils@13.32.2This patch fixes an issue where per-service function configurations were ignored during the build process for experimental V2 services. Consumer groups for queue/v2beta are now correctly scoped by service name to prevent collisions.
vercel-cli(2 releases)
vercel@54.18.1This patch fixes an issue where per-service function configurations were ignored during the build process for experimental V2 services. It also scopes derived queue consumer groups by service name to prevent path collisions.
@vercel/build-utils@13.32.2This patch fixes an issue where per-service function configurations were ignored during the build process for experimental V2 services. Consumer groups for queue/v2beta are now correctly scoped by service name to prevent collisions.
crewai(1 releases)
Track Symbol Changes
Use the Change8 MCP server or GitHub Action to get notified when memory changes.
Learn More