Skip to content

Add cache breakpoints for custom vertex models on Unbound #4585

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

Merged
merged 2 commits into from
Jun 13, 2025
Merged

Add cache breakpoints for custom vertex models on Unbound #4585

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
f38592f
Add cache breakpoints for custom vertex models on Unbound
pugazhendhi-m Jun 12, 2025
6871ae5
Reverts context window
pugazhendhi-m Jun 12, 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
5 changes: 5 additions & 0 deletions src/api/providers/unbound.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9,6 +9,7 @@ import { ApiStream, ApiStreamUsageChunk } from "../transform/stream"
import { convertToOpenAiMessages } from "../transform/openai-format"
import { addCacheBreakpoints as addAnthropicCacheBreakpoints } from "../transform/caching/anthropic"
import { addCacheBreakpoints as addGeminiCacheBreakpoints } from "../transform/caching/gemini"
import { addCacheBreakpoints as addVertexCacheBreakpoints } from "../transform/caching/vertex"

import type { SingleCompletionHandler, ApiHandlerCreateMessageMetadata } from "../index"
import { RouterProvider } from "./router-provider"
Expand Down Expand Up @@ -70,6 +71,10 @@ export class UnboundHandler extends RouterProvider implements SingleCompletionHa
addAnthropicCacheBreakpoints(systemPrompt, openAiMessages)
}
}
// Custom models from Vertex AI (no configuration) need to be handled differently.
if (modelId.startsWith("vertex-ai/google.") || modelId.startsWith("vertex-ai/anthropic.")) {
addVertexCacheBreakpoints(messages)
}

// Required by Anthropic; other providers default to max tokens allowed.
let maxTokens: number | undefined
Expand Down
Loading