Devin AI cd49e66bcc Fix: Allow new Gemini models (gemini-3-pro-preview) to route to native provider ...

- Relaxed Gemini model validation to allow any model starting with gemini-, gemma-, or learnlm- prefixes
- This prevents breakage when new Gemini models are released
- The Google SDK will surface a clear error if a model truly doesn't exist
- Added comprehensive tests for new preview models, case-insensitive validation, and fallback behavior
- Fixes issue #3949

Co-Authored-By: João <joao@crewai.com>

2025-11-19 08:51:35 +00:00

..

agents

fix: ensure instrumentation flags

2025-11-15 20:48:40 -05:00

cassettes

fix: ensure instrumentation flags

2025-11-15 20:48:40 -05:00

cli

feat: fetch and store more data about okta authorization server (#3894)

2025-11-12 15:28:00 -03:00

config

Release/v1.0.0 (#3618)

2025-10-20 14:10:19 -07:00

events

Release/v1.0.0 (#3618)

2025-10-20 14:10:19 -07:00

experimental

chore: don't fail on cleanup error

2025-11-19 01:28:25 -05:00

hooks

Lorenze/feat hooks (#3902)

2025-11-13 10:11:50 -08:00

knowledge

Lorenze/fix duplicating doc ids for knowledge (#3840)

2025-11-06 10:59:52 -08:00

llms

Fix: Allow new Gemini models (gemini-3-pro-preview) to route to native provider

2025-11-19 08:51:35 +00:00

mcp

refactor: improve MCP tool execution handling with concurrent futures (#3854)

2025-11-06 19:28:08 -08:00

memory

Release/v1.0.0 (#3618)

2025-10-20 14:10:19 -07:00

pipeline

Release/v1.0.0 (#3618)

2025-10-20 14:10:19 -07:00

project

fix: hash callback args correctly to ensure caching works

2025-11-05 07:19:09 -05:00

rag

Release/v1.0.0 (#3618)

2025-10-20 14:10:19 -07:00

security

Release/v1.0.0 (#3618)

2025-10-20 14:10:19 -07:00

storage

Release/v1.0.0 (#3618)

2025-10-20 14:10:19 -07:00

telemetry

Release/v1.0.0 (#3618)

2025-10-20 14:10:19 -07:00

tools

Release/v1.0.0 (#3618)

2025-10-20 14:10:19 -07:00

tracing

fix: ensure instrumentation flags

2025-11-15 20:48:40 -05:00

utilities

fix: ensure lite agents course-correct on validation errors

2025-11-05 19:02:11 -05:00

__init__.py

Release/v1.0.0 (#3618)

2025-10-20 14:10:19 -07:00

conftest.py

chore: don't fail on cleanup error

2025-11-19 01:28:25 -05:00

test_context.py

Release/v1.0.0 (#3618)

2025-10-20 14:10:19 -07:00

test_crew_thread_safety.py

Release/v1.0.0 (#3618)

2025-10-20 14:10:19 -07:00

test_crew.py

feat: expose messages to TaskOutput and LiteAgentOutputs (#3880)

2025-11-10 17:38:30 -08:00

test_custom_llm.py

Gl/feat/a2a refactor (#3793)

2025-10-31 18:42:03 -07:00

test_flow_default_override.py

Release/v1.0.0 (#3618)

2025-10-20 14:10:19 -07:00

test_flow_human_input_integration.py

Release/v1.0.0 (#3618)

2025-10-20 14:10:19 -07:00

test_flow_persistence.py

Release/v1.0.0 (#3618)

2025-10-20 14:10:19 -07:00

test_flow_resumability_regression.py

Release/v1.0.0 (#3618)

2025-10-20 14:10:19 -07:00

test_flow_visualization.py

fix: make plot node selection smoother

2025-11-03 07:49:31 -05:00

test_flow.py

fix: handle unpickleable values in flow state

2025-11-06 01:29:21 -05:00

test_hallucination_guardrail.py

Release/v1.0.0 (#3618)

2025-10-20 14:10:19 -07:00

test_imports.py

Release/v1.0.0 (#3618)

2025-10-20 14:10:19 -07:00

test_llm.py

chore: don't fail on cleanup error

2025-11-19 01:28:25 -05:00

test_markdown_task.py

Release/v1.0.0 (#3618)

2025-10-20 14:10:19 -07:00

test_multimodal_validation.py

Release/v1.0.0 (#3618)

2025-10-20 14:10:19 -07:00

test_project.py

chore: don't fail on cleanup error

2025-11-19 01:28:25 -05:00

test_task_guardrails.py

feat: expose messages to TaskOutput and LiteAgentOutputs (#3880)

2025-11-10 17:38:30 -08:00

test_task.py

fix: ensure instrumentation flags

2025-11-15 20:48:40 -05:00

utils.py

Release/v1.0.0 (#3618)

2025-10-20 14:10:19 -07:00