together, standard-tests: specify tool_choice in standard tests #25548
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
|
The latest updates on your projects. Learn more about Vercel for Git ↗︎ 1 Skipped Deployment
|
dosubot
bot
added
size:L
dosubot
bot
added
size:M
baskaryan
reviewed
Aug 19, 2024
| @@ -28,11 +28,22 @@ class TestGroqLlama(BaseTestGroq): | |||
| @property | |||
| def chat_model_params(self) -> dict: | |||
| return { | |||
| "model": "llama-3.1-70b-versatile", | |||
| "model": "llama-3.1-8b-instant", | |||
There was a problem hiding this comment.
ooc why go smaller instead of bigger
There was a problem hiding this comment.
We can do 70b (I changed it from 8b to 70b this morning). My thought is if they support the same features we should prefer smaller models for tests in spirit of testing functionality vs. benchmarking.
| @@ -170,7 +170,14 @@ def test_stop_sequence(self, model: BaseChatModel) -> None: | |||
| def test_tool_calling(self, model: BaseChatModel) -> None: | |||
| if not self.has_tool_calling: | |||
| pytest.skip("Test requires tool calling.") | |||
| model_with_tools = model.bind_tools([magic_function]) | |||
| if self.tool_choice_value == "dict": | |||
There was a problem hiding this comment.
this format is openai-specific (eg anthropic doesn't support i dont think). i actually tihnk more models support just passing in tool name as a string to bind_tools, and think thats what we'd want to standardize on from a devx perspective anyways
There was a problem hiding this comment.
Thanks, updated this to tool name.
baskaryan
approved these changes
Aug 19, 2024
dosubot
bot
added
the
lgtm
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Outdated suggestions cannot be applied.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.
Here we allow standard tests to specify a value for
tool_choicevia atool_choice_valueproperty, which defaults to None.Chat models available in Together have issues passing standard tool calling tests:
Specifying tool_choice also lets us remove an existing
xfailand use a smaller model in Groq tests.