scriptoriumadmin/Roo-Code
1
0
Fork 0
mirror of https://github.com/RooVetGit/Roo-Code.git synced 2026-08-28 05:27:24 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Add cached read and writes to stats and cost calculation for LiteLLM provider (#4206)

Browse source 
* Add cached read and writes to cost calculation for LiteLLM

* Fixed property issue
This commit is contained in:
Mattias Michaux 2025-06-01 22:35:17 +02:00 • committed by GitHub
parent 3fc05adc17
commit dca1076bdb
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
2 changed files with 14 additions and 1 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

2
src/api/providers/fetchers/litellm.ts
View file

@ -58,6 +58,8 @@ export async function getLiteLLMModels(apiKey: string, baseUrl: string): Promise
outputPrice: modelInfo.output_cost_per_token
? modelInfo.output_cost_per_token * 1000000
: undefined,
cacheWritesPrice: modelInfo.cache_creation_input_token_cost ? modelInfo.cache_creation_input_token_cost * 1000000 : undefined,
cacheReadsPrice: modelInfo.cache_read_input_token_cost ? modelInfo.cache_read_input_token_cost * 1000000 : undefined,
description: `${modelName} via LiteLLM proxy`,
}
}

13
src/api/providers/lite-llm.ts
View file

@ -3,6 +3,8 @@ import { Anthropic } from "@anthropic-ai/sdk" // Keep for type usage only
import { litellmDefaultModelId, litellmDefaultModelInfo } from "@roo-code/types"
import { calculateApiCostOpenAI } from "../../shared/cost"
import { ApiHandlerOptions } from "../../shared/api"
import { ApiStream, ApiStreamUsageChunk } from "../transform/stream"
@ -66,7 +68,7 @@ export class LiteLLMHandler extends RouterProvider implements SingleCompletionHa
for await (const chunk of completion) {
const delta = chunk.choices[0]?.delta
const usage = chunk.usage as OpenAI.CompletionUsage
const usage = chunk.usage as LiteLLMUsage
if (delta?.content) {
yield { type: "text", text: delta.content }
@ -82,8 +84,12 @@ export class LiteLLMHandler extends RouterProvider implements SingleCompletionHa
type: "usage",
inputTokens: lastUsage.prompt_tokens || 0,
outputTokens: lastUsage.completion_tokens || 0,
cacheWriteTokens: lastUsage.cache_creation_input_tokens || 0,
cacheReadTokens: lastUsage.prompt_tokens_details?.cached_tokens || 0,
}
usageData.totalCost = calculateApiCostOpenAI(info, usageData.inputTokens, usageData.outputTokens, usageData.cacheWriteTokens, usageData.cacheReadTokens)
yield usageData
}
} catch (error) {
@ -119,3 +125,8 @@ export class LiteLLMHandler extends RouterProvider implements SingleCompletionHa
}
}
}
// LiteLLM usage may include an extra field for Anthropic use cases.
interface LiteLLMUsage extends OpenAI.CompletionUsage {
cache_creation_input_tokens?: number
}