Best memory & reasoning MCP server for ChatGPT (Codex), tested

Dated results

• Memory (reference) passed in ChatGPT (Codex) on 2026-10-20 (Codex CLI 0.162.0-alpha.17.2, @modelcontextprotocol/server-memory 2.0.0, connect time not recorded). Full result
• Sequentialthinking (reference) passed in ChatGPT (Codex) on 2026-10-21 (Codex CLI 0.162.0-alpha.17.2, @modelcontextprotocol/server-sequential-thinking 2.0.0, connect time not recorded). Full result
• Basic Memory passed in ChatGPT (Codex) on 2026-10-22 (Codex CLI 0.162.0-alpha.17.2, basic-memory 0.23.2, connect time not recorded). Full result
• context-mode passed in ChatGPT (Codex) on 2026-10-14 (Codex CLI 0.162.0-alpha.17.2, package version not recorded, connect time not recorded). Full result
• MemPalace passed in ChatGPT (Codex) on 2026-10-22 (Codex CLI 0.162.0-alpha.17.2, mempalace 3.11.0, connect time not recorded). Full result
• Engram passed in ChatGPT (Codex) on 2026-10-21 (Codex CLI 0.162.0-alpha.17.2, server reported v3.5.2, connect time not recorded). Full result

Why Memory (reference) is the candidate

Memory (reference) passed every step in ChatGPT (Codex). It is the group’s site-wide candidate (judged 2026-10-21): Saved a code in one session and recalled it in a fresh session in both Claude Code and Cursor (3 calls in total), using one local JSON file with no network or account. Context-mode also passed recall in both (4 calls) but needs Node 22.5+. Sequential Thinking passed its reasoning task in both, but it has no storage, so it isn't judged on recall. On Oct 14 Memory and context-mode also passed recall in Claude Desktop and ChatGPT (Codex); in Claude Desktop, Memory also listed older codes left in the shared test file by earlier runs. On Oct 19 Memory 2.0.0 (new major, Node 24+) was retested and passed recall again in Claude Code and Cursor. Confirmed Oct 21, 2026: second runs on 2.0.0 passed in all 4 clients (Claude Code and Cursor Oct 19, ChatGPT (Codex) Oct 20, Claude Desktop Oct 21). First runs were on 1.0.0.

Provisional: based on one dated run per app (no repeat runs yet), so treat it as a candidate, not a final pick. How we test.

What failed or is partial in ChatGPT (Codex)

None of these servers failed in ChatGPT (Codex).

Install in ChatGPT (Codex)

The exact snippet we tested is on the Memory (reference) page (pick the ChatGPT (Codex) tab). Not connecting? ChatGPT (Codex) MCP troubleshooting.

Not tested in ChatGPT (Codex) yet

Everything (reference), Supermemory, claude-mem, cognee, agentmemory, Graphiti MCP, MCP Memory Service, Jev Agent, OKF Agent Memory, Jev MCP. Listed from their source; no result in ChatGPT (Codex) yet.

Questions

What is the best memory & reasoning MCP server for ChatGPT (Codex)?

In ToolPassed's hands-on tests in ChatGPT (Codex), Memory (reference) is our provisional candidate for memory & reasoning: it passed install, connect, tools and an agent task on 2026-10-20. We ran 6 memory & reasoning MCP servers in ChatGPT (Codex); 6 passed every step.

Does Memory (reference) work with ChatGPT (Codex)?

Memory (reference) passed in ChatGPT (Codex) on 2026-10-20 (Codex CLI 0.162.0-alpha.17.2, @modelcontextprotocol/server-memory 2.0.0, connect time not recorded).