Weekly Tooling Brief — 2026-10-18

- Added idle compaction in the terminal: a 200K+ token conversation left idle for about 54 minutes is compacted before its one-hour prompt cache expires, so your next message doesn't reprocess it all

* **api:** add regional processing residency options ([#4081](https://github.com/openai/openai-python/issues/4081)) ([21ce0c2](https://github.com/openai/openai-python/commit/21ce0c2eae47b389f25d79095d2ac5aec7812806))

Added native Decisions integration in GPT-Live. Your app can act while your user is speaking or react to what Live says.

If your app registers a global tracer provider, the SDK will create spans with it after upgrading, and they'll be exported with the rest of your traces. Requests will also get trace context headers from the global propagator. To disable, pass `open_telemetry=False` to the client or set `ANTHROPIC_OPEN_TELEMETRY=false`.

Fast mode is out of research preview for Claude Opus 5.5, and you no longer need to request access. Requests to Claude Opus 5.5 no longer require the fast-mode-2026-02-01 beta header, and requests that still send it keep working.

Effective October 14, 2026, at noon PDT (19:00 UTC), GPT-5.5 retires from Codex for users signed in with ChatGPT. The OpenAI API and Codex sessions authenticated with an API key aren’t affected.

* Allow automatic resumption of continuation token for in generate content and generate content stream. Default behavior for this feature is disabled ([83e43ea](https://github.com/googleapis/python-genai/commit/83e43ea082fd101b67a4e61a3c1a12e674881748))

We've launched OpenTelemetry export in beta, so you can trace Messages API requests and Claude Managed Agents sessions in your observability platform. The Claude SDK emits spans in Python, TypeScript, C#, Go, Java, PHP, and Ruby.

- Changed fork subagents (`subagent_type: "fork"`) to be on by default in `claude -p`, the Agent SDK, the desktop app and VS Code, as they already are in the terminal; `CLAUDE_CODE_FORK_SUBAGENT=0` turns them off, and `claude mcp serve` never offers them

Gemini 4 Argon (gemini-4-argon) generally available (GA): Released Gemini 4 Argon (gemini-4-argon), our most capable model built for complex coding, agentic, math, and science tasks.

Added Ultrafast mode to the Agents API. Set agent.service_tier to ultrafast with a supported model to reduce the time between generated tokens.

* **api:** add beta session spending limits ([#4058](https://github.com/openai/openai-python/issues/4058)) ([b570279](https://github.com/openai/openai-python/commit/b5702792a0adb1e86a8b51709e935c1e0acde727))

- Changed local session transcripts to also record each API request's system prompt, tools and parameters, so a transcript you choose to share can reproduce a request exactly

* **api:** add safety identifiers to image requests and responses ([#4093](https://github.com/openai/openai-python/issues/4093)) ([c969e32](https://github.com/openai/openai-python/commit/c969e32ebb325b5910aff4543187fedc5527e172))

* **api:** add Realtime Translation WebSocket sessions ([#4095](https://github.com/openai/openai-python/issues/4095)) ([bfada50](https://github.com/openai/openai-python/commit/bfada5040b0171cd0ae61aefae8c78fcf83c804a))

* **interactions:** Add gemini-4-argon to Model enum ([8554b07](https://github.com/googleapis/python-genai/commit/8554b07d0025e19361b7a10a6384cc143f3a8ea9))

* Concatenate output_text across multi-step model responses in a turn ([180c345](https://github.com/googleapis/python-genai/commit/180c3453f91e4734b828534950fa7f8526934b59))

* **api:** add typed Live sideband events ([#4099](https://github.com/openai/openai-python/issues/4099)) ([8d5cbf0](https://github.com/openai/openai-python/commit/8d5cbf07916b8f2409d940156135ce3c11ac1552))

* add browser and computer toolsets (1/4) ([#4073](https://github.com/openai/openai-python/issues/4073)) ([47b9e4e](https://github.com/openai/openai-python/commit/47b9e4e95e44b1e825dc4e0753967c65d0733445))

* fix(core): serialize file tool operations and make writes atomic (#29078) by @elberthc-byte in https://github.com/google-gemini/gemini-cli/pull/29499