Skip to content

Refactor model cache breakpoint logic to use model info in Unbound #4137

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 1 commit into from
May 30, 2025
Merged

Refactor model cache breakpoint logic to use model info in Unbound #4137

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
11 changes: 8 additions & 3 deletions src/api/providers/unbound.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -5,7 +5,8 @@ import { ApiHandlerOptions, unboundDefaultModelId, unboundDefaultModelInfo } fro

import { ApiStream, ApiStreamUsageChunk } from "../transform/stream"
import { convertToOpenAiMessages } from "../transform/openai-format"
import { addCacheBreakpoints } from "../transform/caching/anthropic"
import { addCacheBreakpoints as addAnthropicCacheBreakpoints } from "../transform/caching/anthropic"
import { addCacheBreakpoints as addGeminiCacheBreakpoints } from "../transform/caching/gemini"

import type { SingleCompletionHandler, ApiHandlerCreateMessageMetadata } from "../index"
import { RouterProvider } from "./router-provider"
Expand Down Expand Up @@ -44,8 +45,12 @@ export class UnboundHandler extends RouterProvider implements SingleCompletionHa
...convertToOpenAiMessages(messages),
]

if (modelId.startsWith("anthropic/claude-3")) {
addCacheBreakpoints(systemPrompt, openAiMessages)
if (info.supportsPromptCache) {
if (modelId.startsWith("google/")) {
addGeminiCacheBreakpoints(systemPrompt, openAiMessages)
} else if (modelId.startsWith("anthropic/")) {
addAnthropicCacheBreakpoints(systemPrompt, openAiMessages)
}
}

// Required by Anthropic; other providers default to max tokens allowed.
Expand Down