-
Notifications
You must be signed in to change notification settings - Fork 2.5k
fix: add missing Vertex AI models and us-south1 region #7726
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.
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
Conversation
- Added gpt-oss-120b-maas model - Added gpt-oss-20b-maas model - Added deepseek-r1-0528-maas model - Added deepseek-v3.1-maas model - Added qwen3-coder-480b-a35b-instruct-maas model - Added qwen3-235b-a22b-instruct-2507-maas model - Added us-south1 to VERTEX_REGIONS for qwen models support Fixes #7725
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Reviewing my own code because apparently I trust no one, not even myself.
Important: The test file webview-ui/src/components/settings/providers/__tests__/Vertex.spec.tsx needs to be updated to include the new us-south1 region in the expected regions array (line 80).
| contextWindow: 128_000, | ||
| supportsImages: false, | ||
| supportsPromptCache: false, | ||
| inputPrice: 2.0, |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Is the decimal format for pricing intentional here? I notice existing models use integers (15, 75) while these new models use decimals (2.0, 8.0). Would it be better to maintain consistency by using 2 instead of 2.0?
| outputPrice: 1.15, | ||
| description: "Meta Llama 4 Maverick 17B Instruct model, 128K context.", | ||
| }, | ||
| "gpt-oss-120b-maas": { |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The "-maas" suffix pattern is interesting. Could we add a comment explaining that these are Model-as-a-Service offerings? This would help future maintainers understand the naming convention.
| { value: "us-west2", label: "us-west2" }, | ||
| { value: "us-west3", label: "us-west3" }, | ||
| { value: "us-west4", label: "us-west4" }, | ||
| { value: "us-south1", label: "us-south1" }, |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Nice attention to detail placing this alphabetically! The ordering is correct between us-west4 and northamerica-northeast1. 👍
Summary
This PR attempts to address Issue #7725 by adding the missing Vertex AI models and the us-south1 region. Feedback and guidance are welcome.
Changes
Added Models
gpt-oss-120b-maas- GPT OSS 120B model via Model-as-a-Servicegpt-oss-20b-maas- GPT OSS 20B model via Model-as-a-Servicedeepseek-r1-0528-maas- DeepSeek R1 0528 model via Model-as-a-Servicedeepseek-v3.1-maas- DeepSeek V3.1 model via Model-as-a-Serviceqwen3-coder-480b-a35b-instruct-maas- Qwen3 Coder 480B Instruct model via Model-as-a-Serviceqwen3-235b-a22b-instruct-2507-maas- Qwen3 235B Instruct model via Model-as-a-ServiceAdded Region
us-south1- Added to VERTEX_REGIONS array for qwen models supportTesting
Related Issue
Fixes #7725
Important
Add missing Vertex AI models and
us-south1region tovertex.ts.gpt-oss-120b-maas,gpt-oss-20b-maas,deepseek-r1-0528-maas,deepseek-v3.1-maas,qwen3-coder-480b-a35b-instruct-maas,qwen3-235b-a22b-instruct-2507-maastovertexModelsinvertex.ts.us-south1toVERTEX_REGIONSinvertex.ts.This description was created by
for 7f08528. You can customize this summary. It will automatically update as commits are pushed.