v1.5.2

1.5.2 (2025-03-25)

Documentation

• update docs to reflect the the addition of Libraries and the Hybrid Knowledge Base type (#1853) (3acaf18)

v1.5.1

1.5.1 (2025-03-24)

Bug Fixes

• demote tokenizer log from warning to debug (#1845) (fc27554)
• deps: bump minimum attrs version to one with attrs.NothingType (#1855) (6e06214)

Documentation

• add docs on openai's reasoning models (#1843) (a4e91e1), closes #1688
• remove duplicate log output (#1841) (da9f9de)

v1.5.0

1.5.0 (2025-03-13)

✨ Features

• configs-drivers-gemini: update default embedding driver model to (9ba2496)
• configs-drivers-gemini: update default prompt driver model to gemini-2.0-flash (9ba2496)
• drivers-embedding-titan: add support for generating image embeddings (1841f0f)
• drivers-embedding-voyageai: add support for generating image embeddings (1841f0f)
• drivers-embedding: add support for embedding ImageArtifacts (1841f0f)
• drivers-local-vector: support persisting multi-modal entries (1841f0f)
• drivers-vector: support upserting/querying with ImageArtifacts (1841f0f)
• drivers: add PerplexityPromptDriver and PerplexityWebSearchDriver (#1755) (7163bf3)
• drivers: update default models to latest for Anthropic/Bedrock Driver Configs (b3e4b68)
• serialization: add serializable_overrides and types_overrides parameters to to_dict/json and from_dict/json methods (c383ba6)
• tasks-prompt: add reflect_on_tool_use flag to disable LLM reflection (#1810) (ecb665d)

🐛 Bug Fixes

• chunker: correctly determine chunk midpoint when empty chunks are present (#1800) (8ec2a8a)
• decorators: preserve type hints on [@observable](https://github.com/observable) decorator (#1767) (0e5a15c)
• drivers-prompt-openai: convert to text rather than error when passing GenericArtifacts (#1830) (fb3f2c4)
• drivers-prompt-openai: remove modalities default (#1774) (7e061b5)
• drivers-vector-marqo: fix upsert failing due to inability to upsert_vectors (#1803) (9b6bfff)
• drivers-vector: don't mutate same instance of meta (#1782) (dd138ca)
• serialization: support serializing more Driver types (c383ba6)

📦 Dependencies

• relax dependencies by removing upper bound constraint (5171332)

📚 Documentation

• add light/dark mode toggle (ff3d2b6)
• add log output to every example that produces it (0407fc7)
• categorize driver nav links (ff3d2b6)
• custom-tools: improve docs for creating custom tools (#1791) (25b1276)
• engines: improve documentation on Rag Engines. (bdc1921)
• improve docs landing page (ff3d2b6)
• improve framework overview (ff3d2b6)
• improve overview page (#1832) (ff3d2b6)
• improve README (ff3d2b6)
• move recipes into framework section (ff3d2b6)
• move recipes into framework section (ff3d2b6)
• recipes: add example of talking to audio file (#1794) (fe27585)
• rename examples to recipes (ff3d2b6)
• serialization: add dedication page on serialization (c383ba6)
• simplify docs landing page (ff3d2b6)
• update contributing guide (ff3d2b6)

v1.4.3

1.4.3 (2025-03-04)

Bug Fixes

• chunker: correctly determine chunk midpoint when empty chunks are present (#1800) (0c1935d)

v1.4.2

1.4.2 (2025-02-28)

Bug Fixes

• drivers-vector: don't mutate same instance of meta (#1782) (bfaaf03)

v1.4.1

1.4.1 (2025-02-27)

Bug Fixes

• decorators: preserve type hints on @observable decorator (#1767) (7001121)
• drivers-prompt-openai: remove modalities default (#1774) (6348c9a)

v1.4.0

1.4.0 (2025-02-24)

Features

• drivers: add GriptapeCloudPromptDriver (#1692) (7af49da)
• drivers: add GrokPromptDriver (f281702)
• drivers: use cloud server events for GriptapeCloudStructureRunDriver and GriptapeCloudAssistantDriver (#1684) (f0fef4c)
• tasks: accept args in BaseTask.run() (#1598) (89c32d4)
• tools: allow setting allowlist and denylist on Tools (#1743) (4ee4515)

Bug Fixes

• drivers-event-listener-griptape-cloud: add type/timestamp fallbacks for custom events (#1758) (ef52194)
• drivers-files-griptape-cloud: pass correct parameters when listing files (#1674) (4b2f618)
• drivers-prompt-azure-openai: fix AzureOpenAiChatPromptDriver by removing unsupported "modalities" (#1694) (6699a3d)
• drivers-prompt-openai: check existence of "audio" attribute (23ca1c5)
• drivers-prompt-openai: conditionally add modalities/reasoning_effort based on model (#1668) (b9311c1)
• drivers: lower max_attempts in ExponentialBackoffMixin to reasonable value (#1725) (29e4575)
• engines-eval: fix evaluation parameters not being included in prompt (#1751) (762958f)
• properly serialize pydantic models (#1757) (ef83084)
• schema: declare all artifacts for deserialization (#1698) (4904309)
• tasks: fix task input being lost on multiple Structure runs (#1732) (f9ac289)
• tasks: remove deprecation warning on CodeExecutionTask.input (#1709) (0427296)
• utils: wrap primitives with more specific artifact types (#1676) (91800ad)

Documentation

• assistants: add section to cloud docs, update nav, add docs pages for what is, getting started, and runs (#1686) (4f6f8ce)
• fix integration test examples (#1724) (4fe4558)
• tasks: improve CodeExecutionTask example (#1710) (8c61d91)
• tools: move tools to framework portion of docs (#1713) (c2b23bb)

v1.3.5

1.3.5 (2025-02-19)

Bug Fixes

• engines-eval: fix evaluation parameters not being included in prompt (#1751) (e267d45)

v1.3.4

1.3.4 (2025-02-13)

Bug Fixes

• drivers: check existence of "audio" attribute (#1730) (db914a3)
• tasks: fix task input being lost on multiple Structure runs (#1732) (f24d5fc)

v1.3.3

1.3.3 (2025-02-12)

Bug Fixes

• tasks: remove deprecation warning on CodeExecutionTask.input (#1709) (a5840b8)

Documentation

• tasks: improve CodeExecutionTask example (#1710) (006e289)