Supermemory changelog, Page 13

Console usage stats, AI SDK prompt templates, and OpenCode updates #

Improvements

You can delete the active consumer organization and its stored data from Profile settings.
The AI SDK now lets developers control how retrieved memories are formatted in model prompts.
Console now shows token and search-query usage for the current and previous billing periods.
OpenCode can recognize custom phrases as requests to save a memory.
OpenCode now loads Supermemory settings that contain trailing commas.

Fixes

MemoryBench no longer assigns the current date when LoCoMo or LongMemEval timestamps cannot be parsed.
Updating a document now processes and bills only newly added content instead of unchanged chunks.

MemoryBench parallel execution and MCP connection reliability #

Improvements

MemoryBench can run benchmark phases concurrently, with separate concurrency controls for ingest, indexing, search, answer, and evaluation.

Fixes

MCP and streaming connections now retry brief server overloads before returning a temporary-unavailable error.

MCP memory limits and Google Drive connector fixes #

Fixes

Memories added through MCP now count against the memory allowance instead of the API token limit.
The Add Memory connection picker now opens Google Drive instead of returning the disabled-connector error.

Google Drive connector fix: connecting from settings restored #

Fixes

Google Drive can be connected again from Settings > Integrations.

OpenCode plugin fixes: memory deletion and context restore #

Fixes

OpenCode's Forget mode now deletes the selected memory correctly.
OpenCode now keeps the active agent when restoring context after compaction.
OpenCode now loads Supermemory correctly when its JSONC configuration contains URLs.
OpenCode setup now disables the current Oh My OpenCode context-recovery hook instead of the obsolete hook name.