Studio: don't apply nest_asyncio on plain CLI starts (breaks asyncio on Python 3.14+)

[NilayYadav]

Problem

run.py calls nest_asyncio.apply() unconditionally at startup. On Python 3.14, asyncio's task tracking moved into C (asyncio.current_task() reads C-level state that nest_asyncio's patched task step never sets), so with the patch applied current_task() returns None inside every request coroutine. anyio then fails on any thread-pool hop — including StaticFiles serving every frontend asset:

File ".../anyio/_backends/_asyncio.py", line 422, in __enter__
    task_state = _task_states[host_task]
File ".../weakref.py", line 326, in __getitem__
    return self.data[ref(key)]
TypeError: cannot create weak reference to 'NoneType' object

Every request 500s; Studio is unusable on 3.14. Minimal repro (Python 3.14.6, anyio 4.13):

import nest_asyncio, asyncio, anyio
nest_asyncio.apply()
async def main():
    print(asyncio.current_task())          # None
    await anyio.to_thread.run_sync(lambda: 1)  # TypeError
asyncio.run(main())

nest_asyncio is archived upstream, so no fix is coming on its side.

Change

Only apply nest_asyncio when the current thread already has a running event loop — the Colab/IPython case it exists for. A plain CLI start has nothing to nest and stays unpatched, which is exactly the environment where Python 3.14 users hit the crash. Notebook behavior on <= 3.13 is unchanged.

Tested

On Python 3.14.6: backend starts, / and /assets/*.js (the previously-crashing StaticFiles thread-pool path) return 200, API routes respond normally. Repro script above confirms the failure mode and that skipping the patch avoids it.

[gemini-code-assist]

Code Review

This pull request modifies studio/backend/run.py to conditionally apply nest_asyncio only when an asyncio event loop is already running. This prevents potential issues on Python 3.14+ where nest_asyncio patches can break asyncio.current_task() during a plain CLI start. There are no review comments, so I have no additional feedback to provide.

Important

The consumer version of Gemini Code Assist on GitHub is being sunset. Starting June 18, 2026, new organization installations will be blocked, and all code review activity will officially cease on July 17, 2026.
For more details on the timeline and next steps, please review the Help Documentation.

[danielhanchen]

@codex review

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 2a61a67ccb

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[chatgpt-codex-connector]

Avoid applying nest_asyncio on Python 3.14 notebooks

When Studio is launched from IPython/Jupyter or another embedder on Python 3.14+, get_running_loop() succeeds, so this branch still calls nest_asyncio.apply(). That patch is global and also affects the uvicorn loop created below in the background thread, so the anyio/thread-pool current_task() failure described in the commit still occurs for notebook/embedded starts even though the package allows Python <3.15. Gate this branch by Python version or avoid patching the server loop.

Useful? React with 👍 / 👎.

[danielhanchen]

Fixed in 5bc376d: gated the apply() on sys.version_info < (3, 14), so nest_asyncio is never touched on 3.14+ (CLI or notebook). Since apply() patches globally it also broke the background uvicorn loop, so notebook/embedded starts 500'd on 3.14 too; skipping it there returns them to 200 while <=3.13 keeps the existing running-loop behavior unchanged. Verified across Python 3.9-3.14.

[danielhanchen]

@codex review

[danielhanchen]

@codex review

[chatgpt-codex-connector]

Codex Review: Didn't find any major issues. Delightful!

Reviewed commit: 5bc376d7d0

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[danielhanchen]

@codex review

[chatgpt-codex-connector]

Codex Review: Didn't find any major issues. 👍

Reviewed commit: 0a2b6b76ce

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[oobabooga]

@danielhanchen LGTM.