Commit Graph
113 Commits
Author SHA1 Message Date
Ankur SharmaandCopybara-Service b2c45f8d91 chore: Enhance the messaging with possible fixes for RESOURCE_EXHAUSTED errors from Gemini
Co-authored-by: Ankur Sharma <ankusharma@google.com>
PiperOrigin-RevId: 833538475
2025-11-17 16:15:56 -08:00
Xuan YangandCopybara-Service 1dd97f5b45 fix: Add experimental feature to use parameters_json_schema and response_json_schema for McpTool
Co-authored-by: Xuan Yang <xygoogle@google.com>
PiperOrigin-RevId: 833144947
2025-11-16 20:55:16 -08:00
George WealeandCopybara-Service 22eb7e5b06 feat: Add support for parsing inline JSON tool calls in LiteLLM responses
Close #1968

Co-authored-by: George Weale <gweale@google.com>
PiperOrigin-RevId: 831911719
2025-11-13 10:18:27 -08:00
Liang WuandCopybara-Service 22c6dbe83c chore: Defer import of live, Client and _transformers in google.genai
Co-authored-by: Liang Wu <wuliang@google.com>
PiperOrigin-RevId: 831444358
2025-11-12 10:27:56 -08:00
George WealeandCopybara-Service a19be12c1f fix: change LiteLLM content and tool parameter handling
This changes how content parts are converted for LiteLLM, treating all "text/" mime types as plain text and only "application/pdf" as a file

Close #1940

Co-authored-by: George Weale <gweale@google.com>
PiperOrigin-RevId: 831418696
2025-11-12 09:23:00 -08:00
Google Team MemberandCopybara-Service b2c8ba5806 chore: Defer import of live, Client and _transformers in google.genai
This is targeting the Client import mostly, but also prevents future latency increase if the other modules in genai adds more 3p dependencies. This change will make ADK only import `live`, `Client` and `_transformers` just-in-time, therefore cutting down cold start latency.

PiperOrigin-RevId: 831244349
2025-11-11 23:56:20 -08:00
Liang WuandCopybara-Service 22ca7eefc0 chore: Defer import of live, Client and _transformers in google.genai
This is targeting the Client import mostly, but also prevents future latency increase if the other modules in genai adds more 3p dependencies. This change will make ADK only import `live`, `Client` and `_transformers` just-in-time, therefore cutting down cold start latency.

Co-authored-by: Liang Wu <wuliang@google.com>
PiperOrigin-RevId: 831124329
2025-11-11 17:01:28 -08:00
George WealeandCopybara-Service 93aad61198 fix: Safely handle FunctionDeclaration without a required attribute
Update `_function_declaration_to_tool_param` to use `getattr` when accessing the `required` field within `function_declaration.parameters`. This prevents `AttributeError` when a `FunctionDeclaration`'s parameters schema does not include a `required` attribute
Close #2891

Co-authored-by: George Weale <gweale@google.com>
PiperOrigin-RevId: 830225582
2025-11-09 19:06:45 -08:00
Sanjay Siddharth MJandCopybara-Service 2882995289 fix: Fixes DeprecationWarning when using send method
Merge https://github.com/google/adk-python/pull/3352

Replace send() method with send_realtime_input() to fix DeprecationWarning

### Link to Issue or Description of Change

**1. Link to an existing issue (if applicable):**

- Closes: #2393

### Testing Plan

Have run unit test for test_live_request_queue.py

**Unit Tests:**

- [ ] I have added or updated unit tests for my change.
- [x] All unit tests pass locally.

Summary :

tests/unittests/agents/test_live_request_queue.py::test_send_realtime PASSED                [100%]

**Manual End-to-End (E2E) Tests:**

### Checklist

- [x] I have read the [CONTRIBUTING.md](https://github.com/google/adk-python/blob/main/CONTRIBUTING.md) document.
- [x] I have performed a self-review of my own code.
- [x] I have commented my code, particularly in hard-to-understand areas.
- [ ] I have added tests that prove my fix is effective or that my feature works.
- [x] New and existing unit tests pass locally with my changes.
- [x] I have manually tested my changes end-to-end.
- [ ] Any dependent changes have been merged and published in downstream modules.

Tested both API variants.

Co-authored-by: Hangfei Lin <hangfei@google.com>
COPYBARA_INTEGRATE_REVIEW=https://github.com/google/adk-python/pull/3352 from SanjaySiddharth:fix-dep-session.send b23b641a63ff20d1f2dbd9abd519f8facf0aab77
PiperOrigin-RevId: 830182844
2025-11-09 16:00:43 -08:00
Google Team MemberandCopybara-Service 34ea2edfd2 fix: Fixes a bug introduced by the underlying google genai library
We need to set project and location parameters when we instantiate the Google GenAI library Client for VertexAi. This used to work before through environment variables but it doesn't seem to work anymore. So we are updating the `ApigeeLlm` wrapper to read from `GOOGLE_CLOUD_PROJECT` and `GOOGLE_CLOUD_LOCATION` environment variables and pass to the Client constructor.

PiperOrigin-RevId: 829569362
2025-11-07 14:02:41 -08:00
Google Team MemberandCopybara-Service f1f44675e4 ADK changes
PiperOrigin-RevId: 829136628
2025-11-06 15:55:26 -08:00
George WealeandCopybara-Service 7ea4aed35b fix: Add support for structured output schemas in LiteLLM models
Add `_to_litellm_response_format` to convert ADK's `response_schema` types (Pydantic models, JSON schema dicts) into the format needed by LiteLLM for JSON object/schema constraints

Close #1967

Co-authored-by: George Weale <gweale@google.com>
PiperOrigin-RevId: 829037987
2025-11-06 11:29:10 -08:00
Hangfei LinandCopybara-Service 1819ecb4b8 fix: Improve handling of partial and complete transcriptions in live calls
In `gemini_llm_connection.py`, accumulate partial transcription texts and emit `LlmResponse` with `partial=True` for each chunk. When the transcription is marked as `finished`, emit a final `LlmResponse` with the full accumulated text and `partial=False`.

In `runners.py`, modify `_should_append_to_history` to only add transcription events to the history when they are fully finished, preventing partial transcriptions from being added.

Co-authored-by: Hangfei Lin <hangfei@google.com>
PiperOrigin-RevId: 829029715
2025-11-06 11:09:21 -08:00
Krzysztof CzuszynskiandCopybara-Service 8dd5a79b29 fix: Fix transcript finish
Merge https://github.com/google/adk-python/pull/3324

**Problem:**
ADK seems to not pass output/input transcription `finished` flag from the Gemini.

**Solution:**
Relaxation of checking conditions of valid llm_response for input/output transcription.

### Testing Plan

Unit test `test_receive_transcript_finished` checks if input/output transcription message with no text but `finished` flag is received.

**Unit Tests:**

- [x] I have added or updated unit tests for my change.
- [x] All unit tests pass locally.

<img width="785" height="373" alt="image" src="https://github.com/user-attachments/assets/6d870e9f-1372-4808-91a9-38578c1b3729" />

**Manual End-to-End (E2E) Tests:**

Configure Gemini Agent to produce input & output transcriptions. Observe incoming transcript messages - to see if the finished flag appears at the end agents & users statement.

### Checklist

- [x] I have read the [CONTRIBUTING.md](https://github.com/google/adk-python/blob/main/CONTRIBUTING.md) document.
- [x] I have performed a self-review of my own code.
- [x] I have commented my code, particularly in hard-to-understand areas.
- [x] I have added tests that prove my fix is effective or that my feature works.
- [x] New and existing unit tests pass locally with my changes.
- [x] I have manually tested my changes end-to-end.
- [x] Any dependent changes have been merged and published in downstream modules.

### Additional context

The mentioned `finished` flag can be obtained in native Gemini APIs like via Websocket but not via ADK.

Co-authored-by: Hangfei Lin <hangfei@google.com>
COPYBARA_INTEGRATE_REVIEW=https://github.com/google/adk-python/pull/3324 from ChrisQlasty:fix/transcript_finish e7b8e5e5f5fac1d2dfd495f2dbadb19ed4328b0c
PiperOrigin-RevId: 828987996
2025-11-06 09:29:12 -08:00
Kathy WuandCopybara-Service 6e5c0eb6e0 feat: Add token usage to live events for bidi streaming
Populate the usage_metadata field for live events with the metadata provided by the Gemini live API.

Co-authored-by: Kathy Wu <wukathy@google.com>
PiperOrigin-RevId: 828124232
2025-11-04 14:01:25 -08:00
Eliza HuangandCopybara-Service 4f85e86fc3 feat: Add support for extracting cache-related token counts from LiteLLM usage
Closes #3049

Co-authored-by: Eliza Huang <heliza@google.com>
PiperOrigin-RevId: 828091671
2025-11-04 12:43:06 -08:00
George WealeandCopybara-Service ce91a8ef73 fix: Pass drop_params to LiteLLM completion API
This lets users to specify `drop_params` when initializing `LiteLlm`, which will be forwarded to LiteLLM's `acompletion` or `completion` calls

Close #1718

Co-authored-by: George Weale <gweale@google.com>
PiperOrigin-RevId: 828058105
2025-11-04 11:19:58 -08:00
Shan CaoandCopybara-Service d4c63fc562 chore: Add model tracking to LiteLlm and introduce a LiteLLM with fallbacks demo
Related: #2292

Co-authored-by: Shan Cao <caoshan@google.com>
PiperOrigin-RevId: 828024955
2025-11-04 10:10:09 -08:00
George WealeandCopybara-Service 489c39db01 fix: Remove redundant format field from LiteLLM content objects
LiteLLM providers can extract the MIME type from the data URI. Removing the separate `format` field avoids redundancy and potential issues with backends that may reject requests containing this field.

Close #2017

Co-authored-by: George Weale <gweale@google.com>
PiperOrigin-RevId: 828014286
2025-11-04 09:44:38 -08:00
ce8f674a28 fix: Allow LLM request to override the model used in the generate content async method in LiteLLM
Merge https://github.com/google/adk-python/pull/3066

Close #3065

Co-authored-by: Raman Mangla <ramanmangla@google.com>
PiperOrigin-RevId: 825880794
2025-10-29 23:34:33 -07:00
Eliza HuangandCopybara-Service f9569bbb1a fix: Enable usage metadata in LiteLLM streaming
Closes #3181

Co-authored-by: Eliza Huang <heliza@google.com>
PiperOrigin-RevId: 825833660
2025-10-29 20:37:54 -07:00
Eliza HuangandCopybara-Service 71aa5645f6 fix: Propagate LiteLLM finish_reason to LlmResponse for use in callbacks
Closes #3114.

Co-authored-by: Eliza Huang <heliza@google.com>
COPYBARA_INTEGRATE_REVIEW=https://github.com/google/adk-python/pull/3319 from lizzij:fix/litellm-finish-reason-3109 da6ed0aec514231fc274fee744f885ae95b9d45c
PiperOrigin-RevId: 825783229
2025-10-29 17:55:03 -07:00
Ayush AgrawalandCopybara-Service 2f72ceb49b chore: Do not treat FinishReason.STOP as error case for LLM responses containing candidates with empty contents
Fixes #2905

PiperOrigin-RevId: 825263302
2025-10-28 16:54:50 -07:00
George WealeandCopybara-Service 74d8361a7e fix: Add a fallback user message to LiteLLM requests if the last user message is empty
Related to #3255
Close #2560

PiperOrigin-RevId: 825143315
2025-10-28 11:56:47 -07:00
Google Team MemberandCopybara-Service 87dcb3f7ba feat: Add ApigeeLlm as a model that let's ADK Agent developers to connect with an Apigee proxy
PiperOrigin-RevId: 824712152
2025-10-27 15:53:33 -07:00