build(deps): bump pydantic-ai from 2.43.0 to 2.51.0

[dependabot]

Bumps pydantic-ai from 2.43.0 to 2.51.0.

Release notes

Sourced from pydantic-ai's releases.

v2.51.0 (2026-09-25)

What's Changed

⚠️ Compatibility Notes

• Match dated gemini-3.8-live ids, reject google_affective_dialog on Gemini 3.1 Flash Live and 3.8 Live at connect, and raise RealtimeError for Gemini Live close codes by @​DouweM in pydantic/pydantic-ai#8761
• Raise a UserError for a realtime tool_choice that forces a tool call on OpenAI, Azure OpenAI, and xAI by @​DouweM in pydantic/pydantic-ai#8755

🚀 Features

• Add OpenAI GPT-Live support with OpenAILiveModel by @​DouweM in pydantic/pydantic-ai#8390
• Expose context_window_used on realtime sessions, from GPT-Live's reported ratio or response usage by @​DouweM in pydantic/pydantic-ai#8803

🐛 Bug Fixes

• Reduce RunContext copying overhead in capability hooks by @​Kludex in pydantic/pydantic-ai#8775
• Keep provider_details on OpenAI Realtime tool-call responses and the response id on replies cut off by a drop or close() by @​DouweM in pydantic/pydantic-ai#8762
• Buffer text deltas in CompletedStreamedResponse replay by @​Kludex in pydantic/pydantic-ai#8799
• Resend a recent image with a typed Gemini Live turn, which doesn't see video frames by @​DouweM in pydantic/pydantic-ai#8756
• Cache the agent graph instead of rebuilding it on every Agent.run() by @​Kludex in pydantic/pydantic-ai#8658
• Stop spawning a task group when a toolset or capability fan-out has a single child by @​Kludex in pydantic/pydantic-ai#8774

Full Changelog: pydantic/pydantic-ai@v2.50.0...v2.51.0

v2.50.0 (2026-09-24)

What's Changed

⚠️ Compatibility Notes

• Ask a DecisionModel which route to take by name, with one label per route by @​DouweM in pydantic/pydantic-ai#8733
• Make ModelSelectionContext.messages end with the request being routed, and add ModelSelectionContext.prompt by @​DouweM in pydantic/pydantic-ai#8737
• Replace the decision model's tool-call lean with an opt-in decision_route_threshold by @​DouweM in pydantic/pydantic-ai#8739

🚀 Features

• Expose OpenAI service_tier in provider details by @​pydanty in pydantic/pydantic-ai#7945
• Add DecisionModel, a base for Decisions-protocol models, and make TypeSafeModel one by @​DouweM in pydantic/pydantic-ai#8696
• Add RunContext.in_durable_context so hooks can tell they run in durable workflow code by @​mpfaffenberger in pydantic/pydantic-ai#8723
• Add gemini-3.8-live and gemini-3.8-live-extended-thinking realtime support by @​DouweM in pydantic/pydantic-ai#8393
• Ask a DecisionModel's route fields with the route question under a route premise, with nested-field context and a done state after tool calls by @​DouweM in pydantic/pydantic-ai#8749
• Emit a decide span for every request a DecisionModel sends by @​DouweM in pydantic/pydantic-ai#8698

🐛 Bug Fixes

• Keep the type variable of a generic OutputSpec[T] passed to Agent.run under mypy by @​DouweM in pydantic/pydantic-ai#8730
• Send ThinkingParts in the history decision models judge by @​DouweM in pydantic/pydantic-ai#8738
• Price Anthropic one-hour cache writes at the one-hour rate instead of the five-minute rate by @​kazmer97 in pydantic/pydantic-ai#8666
• Price realtime image input at the image rate by @​DouweM in pydantic/pydantic-ai#8741
• Add RequestUsage.audio_seconds so duration-billed voice models are priced by @​DouweM in pydantic/pydantic-ai#8427
• Offer a decision model's unfillable output_type as a route beside other routes, and rename ToolCallProposed to UnfillableRoute under a DecisionHandOff base by @​DouweM in pydantic/pydantic-ai#8744
• Keep realtime user turns whole when the microphone streams continuously on providers without input item IDs by @​DouweM in pydantic/pydantic-ai#8746
• Roll back a realtime send the provider refuses, so wait_for_reply() no longer hangs on it by @​DouweM in pydantic/pydantic-ai#8747
• Turn reasoning off for thinking=False on OpenAI Realtime by sending reasoning.effort: 'none' by @​DouweM in pydantic/pydantic-ai#8753
• Take Jev's context_window from genai-prices and document compacting a Jev conversation by @​DouweM in pydantic/pydantic-ai#8740
• Buffer five minutes of audio per stream_audio() view so speaker-paced playback gets whole replies by @​DouweM in pydantic/pydantic-ai#8750
• Stop Google Search from crashing Gemini 2.5 native-audio realtime sessions by @​DouweM in pydantic/pydantic-ai#8751
• Keep user speech out of Gemini 3.x realtime history when input transcription is off by @​DouweM in pydantic/pydantic-ai#8754

... (truncated)

Commits

• 39328d5 Resend a recent image with a typed Gemini Live turn, which doesn't see video ...
• f13ba97 Buffer text deltas in CompletedStreamedResponse replay (#8799)
• bc852ab Expose context_window_used on realtime sessions, from GPT-Live's reported r...
• 0cf5bf5 Replay recorded WebSocket timing so the GPT-Live continuous-microphone parity...
• 656e82a Keep provider_details on OpenAI Realtime tool-call responses and the respon...
• d0ea063 Stop spawning a task group when a toolset or capability fan-out has a single ...
• 6eab793 Add OpenAI GPT-Live support with OpenAILiveModel (#8390)
• 10bc177 Raise a UserError for a realtime tool_choice that forces a tool call on O...
• 1aff77c Match dated gemini-3.8-live ids, reject google_affective_dialog on Gemini...
• 694a080 Use instructions instead of system_prompt in docs examples that aren't ab...
• Additional commits viewable in compare view

[dependabot]

Labels

The following labels could not be found: python. Please create it before Dependabot can add it to a pull request.

Please fix the above issues or remove invalid values from dependabot.yml.

[fmueller]

@dependabot rebase