-
Couldn't load subscription status.
- Fork 93
Support concurrent tool call streaming with ID-based tracking #1468
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
base: 10-24-add_unittest_for_mirrored_provider_clients_anthropicvertex_
Are you sure you want to change the base?
Conversation
Deploying with
|
| Status | Name | Latest Commit | Updated (UTC) |
|---|---|---|---|
| ✅ Deployment successful! View logs |
v2-docs | baa82ae | Oct 29 2025, 02:38 PM |
|
Warning This pull request is not mergeable via GitHub because a downstack PR is open. Once all requirements are satisfied, merge this PR as a stack on Graphite.
This stack of pull requests is managed by Graphite. Learn more about stacking. |
3ba3761 to
7c91809
Compare
ec1ba56 to
02e948c
Compare
7c91809 to
f0426a6
Compare
f0426a6 to
baa82ae
Compare

TL;DR
Added support for parallel tool calls in streaming responses by tracking tool calls by ID.
What changed?
idfield toToolCallChunkandToolCallEndChunkclasses to uniquely identify tool calls_current_tool_calls_by_id)How to test?
Why make this change?
This change enables proper handling of parallel tool calls in streaming responses. By tracking tool calls by their unique IDs rather than using a single
_current_contentreference, the system can now process multiple tool calls simultaneously without losing track of their state. This is particularly important for complex interactions where multiple tools might be called in parallel by the LLM.