Who contributed to OpenAI on May 27, 2026?

4 developers shipped this update, including rhan-oai, canvrno-oai, jif-oai, and romainhuet.

What were the notable OpenAI updates?

Restore legacy image detail values, Move slash input logic out of chat composer, and Move MCP tool naming mode into manager (#21576).

OpenAI

Codex, the SDKs, and the engine behind ChatGPT

Pick a date

Topics: AI / ML Full archive →

The Wire · Showcase

CODEX STABILIZES LEGACY CHECKPOINTS WHILE PYTHON SDK SHIFTS TO GPT-5.5

By RepoJournal · Filed 06:03 UTC on May 27, 2026 · About OpenAI

4 people shipped this

rhan-oai @rhan-oai 1 cited

canvrno-oai @canvrno-oai 1 cited

jif-oai @jif-oai 1 cited

romainhuet @romainhuet 1 cited

Codex shipped a compatibility fix for older persisted rollouts that were breaking resume workflows, while the Python SDK quietly updated all README examples to point at the next-generation models.

The Codex team fixed a deserialization issue where legacy `input_image.detail` values from before the enum was narrowed were causing resume to skip checkpoints and reconstruct oversized suffixes [1]. This fix addresses confirmed Sentry reports and prevents compaction failures in production. In parallel, Codex continues its modularization push: slash-command logic moved out of the chat composer to prevent file bloat [2], MCP tool naming is now centralized in the manager to keep `ToolInfo` as the single source of truth [3], and goal telemetry parity was added as the `/goal` endpoint migrates into the extension system to preserve existing metric contracts [4]. The Python SDK updated all README snippets to reference `gpt-5.5` and `gpt-realtime-2` [6], signaling these are now the canonical examples developers should reference. Codex also added experimental `additionalContext` support to turn endpoints, allowing clients to pass ephemeral external state without surfacing it in prompts or triggering lifecycle behavior [5].

Action items

→ Verify your Codex deployment includes the legacy detail values fix to prevent resume failures openai/codex [immediate]
→ Update any internal documentation or examples still pointing to older GPT models toward gpt-5.5 openai/openai-python [plan]
→ Test additionalContext parameter if you're building client-side integrations with Codex turn endpoints openai/codex [monitor]

References

[1] Restore legacy image detail values ↗ openai/codex
[2] Move slash input logic out of chat composer ↗ openai/codex
[3] Move MCP tool naming mode into manager (#21576) openai/codex
[4] Add goal extension telemetry parity ↗ openai/codex
[5] Add experimental turn additional context (#24154) openai/codex
[6] Update README models to gpt-5.5 and gpt-realtime-2 ↗ openai/openai-python

Quick answers

What shipped in OpenAI on May 27, 2026?: Codex shipped a compatibility fix for older persisted rollouts that were breaking resume workflows, while the Python SDK quietly updated all README examples to point at the next-generation models. In total, 48 commits, 48 pull requests, and 1 releases landed.
Who contributed to OpenAI on May 27, 2026?: 4 developers shipped this update, including rhan-oai, canvrno-oai, jif-oai, and romainhuet.
What were the notable OpenAI updates?: Restore legacy image detail values, Move slash input logic out of chat composer, and Move MCP tool naming mode into manager (#21576).

CODEX HARDENS MCP RUNTIME STABILITY ACROSS FIVE CRITICAL FIXES

OpenAI shipped a coordinated hardening of the Model Context Protocol layer across Codex, patching race conditions in thread startup, resource access, and authority management that could leave stale configurations in production.

ai 66 shipped 1-min read

OpenAI 1 day ago

SDK RELEASES LOCK IN PROMPT CACHING AND SPEND CONTROLS

Both Node and Python clients shipped v6.49.0 and v2.48.0 overnight with admin spend limit APIs and relaxed prompt cache validation.

ai 103 shipped 1-min read

OpenAI 2 days ago

CODEX CONSOLIDATES THREAD STARTUP, NODE SDK ADDS REALTIME AND STANDARD SCHEMA SUPPORT

Codex refactored its threading and database layer for consistency while the Node SDK ships four new schema helpers and fixes audio streaming in a single release push.

ai 120 shipped 1-min read

OpenAI 3 days ago

CODEX FIXES WINDOWS EXECUTION BUGS WHILE TIGHTENING SECURITY REVIEW GATES

Codex shipped critical fixes for Windows process management and terminal navigation that were silently breaking execution sessions on Windows machines.

ai 104 shipped 1-min read

Elsewhere on the wire

AI Agents about 9 hours ago

CLAUDE OPUS 5 LANDS ACROSS THE STACK

The newest Anthropic model is now live in langchain, Cline, and llama-index, with native support for extended reasoning and 1M context windows.

ai-agents 28 shipped 1-min read

Local LLMs about 9 hours ago

OLLAMA LANDS LAGUNA SUPPORT AND CRUSHES MEMORY LEAKS WHILE SGLANG HITS V0.5.16 WITH CONFIDENCE-DRIVEN SPECULATIVE DECODING

Ollama shipped three critical performance and reliability fixes for Metal residency and concurrent access patterns, while SGL-Lang released 0.5.16 with a new speculative algorithm hitting 383.7 tok/s on DeepSeek-V4.