HTTP request bodies now limited in size in version 1.30.1.
Version 1.30.1 limits HTTP request sizes and JSON-RPC batch lengths.
Move between the fast signal, the full context and the practical next move without losing the thread.
Models, agents, robotics, research, business and policy, with the original source attached. Search and move beyond the latest thirty updates.
Open the 30-day AI Change IndexVersion 1.30.1 limits HTTP request sizes and JSON-RPC batch lengths.
The new version of @modelcontextprotocol/node adds OAuth scope challenges for various tools and resources. It also enforces a 4 MiB default size limit on HTTP request bodies.
The codemod scanner now correctly identifies module-specifier positions, preventing misclassification of client-only projects.
v3.0.11 adds MMR reranker for diverse search results and RecencyReranker for prioritizing newer content.
v4.69.0 includes a new GPT-Rosalind research model. It also clarifies Chat Completions seed bounds.
GCP external storage support added in the 2026-09-23 release.
The new GCP external storage support is now available in version 3.66.0.
v7.23.0 adds GCP external storage support and a GPT-Rosalind research model.
Version 2.48.0 of the software now supports OpenAI's gpt-6-sol and gpt-6-luna models.
The 2026-09-23 release includes voice resources for Java and SpeechMetadata in GenAI SDKs.
Version 3.19.0 includes GCP external storage support and a new GPT-Rosalind research model.
Sub-agents now run tool calls simultaneously, and models with large output limits receive a bigger default budget.
The new version 4.68.0 adds GPT-6 model identifiers. It also fixes stream preservation during multipart uploads.
The v0.94.0 release includes GPT-6 model identifiers for the API. It also updates development dependencies.
GPT-6 model identifiers for Sol and Luna added in v3.65.0.
The 7.22.0 release includes identifiers for GPT-6 Sol and Luna models.
New GPT-6 model identifiers for Sol and Luna added in v3.18.0.
The 2.24.0 release includes new Voice features and bug fixes in the GAOS SDK.
v2.47.0: TypeSafeModel now supports three additional field types.
Version 2.25.0 of the GAOS SDK now includes sample_audio and a Voices API resource.
The 4.67.0 release includes session environment reset events for the API. It also clarifies external storage deletion behavior.
The 2026-09-22 release of v3.64.3 includes a fix to preserve legacy model choices in the API.
The 7.21.0 release includes a new API feature for session environment reset events. It also fixes a bug to preserve model choices and improve request handling.
The update allows system messages to be placed anywhere in the conversation. Multiple leading system messages are now merged instead of rejected.