feat: Tool-Use functions call #4
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
## [0.7.1-rc.1](0.7.0...0.7.1-rc.1) (2025-02-08) ### ♻️ Improvements * Remove tool_calls from Message struct and update related tests ([265bdec](265bdec)) ### 👷 CI * Improve the Release Process Allowing to have Release Candidates ([#3](#3)) ([b97e5da](b97e5da)) ### 📚 Documentation * Remove unnecessary code block from README.md ([dcb7c26](dcb7c26))
…rameters updates, use JSON raw value for now Signed-off-by: Eden Reich <[email protected]>
## [0.7.1-rc.2](0.7.1-rc.1...0.7.1-rc.2) (2025-02-09) ### 🐛 Bug Fixes * Enhance Tool and ToolFunction structures with description and parameters updates, use JSON raw value for now ([cdf6eca](cdf6eca))
|
🎉 This PR is included in version 0.7.1-rc.2 🎉 The release is available on GitHub release Your semantic-release bot 📦🚀 |
…ify code Will be using regular JSON for this part. Signed-off-by: Eden Reich <[email protected]>
…update ToolCallResponse structure Signed-off-by: Eden Reich <[email protected]>
## [0.7.1-rc.3](0.7.1-rc.2...0.7.1-rc.3) (2025-02-09) ### ♻️ Improvements * Remove ToolParameterType and ToolParameter structs to simplify code ([0e4d1b4](0e4d1b4)) * Rename parameters to arguments in ToolFunctionResponse and update ToolCallResponse structure ([3f36c5f](3f36c5f))
|
🎉 This PR is included in version 0.7.1-rc.3 🎉 The release is available on GitHub release Your semantic-release bot 📦🚀 |
…ation tests Signed-off-by: Eden Reich <[email protected]>
## [0.8.0-rc.1](0.7.1-rc.3...0.8.0-rc.1) (2025-02-09) ### ✨ Features * Add optional tool_call_id to Message struct and update serialization tests ([d51e911](d51e911))
|
🎉 This PR is included in version 0.8.0-rc.1 🎉 The release is available on GitHub release Your semantic-release bot 📦🚀 |
Signed-off-by: Eden Reich <[email protected]>
Signed-off-by: Eden Reich <[email protected]>
|
Thinking if the API is good, I might want to change it to be something like: client.with_tools(Some(tools)).generate_content(provider, model, messages);Instead of: client.generate_content(provider, model, messages, tools);Also the messages, I want to avoid having to add tool_call_id to all messages, this should be optional so probably I need to introduce two type of messages with default values. The benefits are separation of concern - the function generate_content doesn't need to change if the tools are passed or not. |
…dling and return details Signed-off-by: Eden Reich <[email protected]>
Signed-off-by: Eden Reich <[email protected]>
This way I don't need to reference None in cases where tools is not needed. It simplifies the usage of generate_content function. Signed-off-by: Eden Reich <[email protected]>
This makes the code looks more readable, devs should not be concerns with tool_call_id by setting it to None if they not necessarily using tools, it's only valid for MessageRole of type Tool. Signed-off-by: Eden Reich <[email protected]>
…ssible with_tools could be defined on initialization for all subsequent generate_content requests or for a particular request. Signed-off-by: Eden Reich <[email protected]>
Signed-off-by: Eden Reich <[email protected]>
… it at the bottom Signed-off-by: Eden Reich <[email protected]>
…ange Signed-off-by: Eden Reich <[email protected]>
## [0.8.0-rc.2](0.8.0-rc.1...0.8.0-rc.2) (2025-02-10) ### ♻️ Improvements * Implement Default trait for Message and MessageRole structs ([cb43ca8](cb43ca8)) * Initialize the client with tools optionally ([9c56170](9c56170)) * Use in the tests two different conventions to see what's possible ([eef16e8](eef16e8)) ### 📚 Documentation * Enhance InferenceGatewayAPI docblocks with additional error handling and return details ([b666c33](b666c33)) * Improve the docs for Tool-Use, update it ([9e5f5da](9e5f5da)) * Improve the example by mentioning the workflow of messages exchange ([ef474b4](ef474b4)) * Remove redundant whitespace in InferenceGatewayAPI documentation ([422e857](422e857)) * Resort functions examples, healthcheck is less important so put it at the bottom ([e6eb55d](e6eb55d)) * Update the example in the README.md ([0fda0dc](0fda0dc)) ### ✅ Miscellaneous * Add assertion for tool_calls length in response test ([4edcd4b](4edcd4b))
|
🎉 This PR is included in version 0.8.0-rc.2 🎉 The release is available on GitHub release Your semantic-release bot 📦🚀 |
github-actions bot
pushed a commit
that referenced
this pull request
Feb 10, 2025
## [0.8.0](0.7.0...0.8.0) (2025-02-10) ### ✨ Features * Tool-Use functions call ([#4](#4)) ([efd8e6b](efd8e6b)), closes [#3](#3) ### ♻️ Improvements * Remove tool_calls from Message struct and update related tests ([265bdec](265bdec)) ### 👷 CI * Improve the Release Process Allowing to have Release Candidates ([#3](#3)) ([b97e5da](b97e5da)) ### 📚 Documentation * Remove unnecessary code block from README.md ([dcb7c26](dcb7c26))
|
🎉 This PR is included in version 0.8.0 🎉 The release is available on GitHub release Your semantic-release bot 📦🚀 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Summary
This PR should improve the usage of the LLM Tools.