Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

feat: improve reliability for vertex models by setting ANY mode #2437

Merged
merged 4 commits into from
Feb 15, 2025
Merged

feat: improve reliability for vertex models by setting ANY mode #2437

Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
745c78a
feat: Fix VLLM tool_choice (#1001)
mattzh72 Feb 14, 2025
1c48924
chore: Feb 14th Sync (#1007)
mattzh72 Feb 14, 2025
1bf26d4
feat: use function calling mode `ANY` for vertex (#1008)
sarahwooders Feb 15, 2025
ca245ea
Merge branch 'main' into add-vertext-tool-any
sarahwooders Feb 15, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
16 changes: 14 additions & 2 deletions letta/llm_api/google_vertex.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -307,15 +307,27 @@ def google_vertex_chat_completions_request(
"""

from google import genai
from google.genai.types import FunctionCallingConfig, FunctionCallingConfigMode, ToolConfig

client = genai.Client(vertexai=True, project=project_id, location=region, http_options={"api_version": "v1"})
# add dummy model messages to the end of the input
if add_postfunc_model_messages:
contents = add_dummy_model_messages(contents)

tool_config = ToolConfig(
function_calling_config=FunctionCallingConfig(
# ANY mode forces the model to predict only function calls
mode=FunctionCallingConfigMode.ANY,
)
)
config["tool_config"] = tool_config.model_dump()

# make request to client
response = client.models.generate_content(model=model, contents=contents, config=config)
print(response)
response = client.models.generate_content(
model=model,
contents=contents,
config=config,
)

# convert back response
try:
Expand Down
Loading