Skip to content

feat: update Gemini and Vertex AI models from preview to GA version #5907

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

Closed
wants to merge 1 commit into from
Closed

feat: update Gemini and Vertex AI models from preview to GA version #5907

Show file tree
Hide file tree
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
74 changes: 0 additions & 74 deletions packages/types/src/providers/gemini.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -68,80 +68,6 @@ export const geminiModels = {
inputPrice: 0,
outputPrice: 0,
},
"gemini-2.5-pro-preview-03-25": {
maxTokens: 65_535,
contextWindow: 1_048_576,
supportsImages: true,
supportsPromptCache: true,
inputPrice: 2.5, // This is the pricing for prompts above 200k tokens.
outputPrice: 15,
cacheReadsPrice: 0.625,
cacheWritesPrice: 4.5,
tiers: [
{
contextWindow: 200_000,
inputPrice: 1.25,
outputPrice: 10,
cacheReadsPrice: 0.31,
},
{
contextWindow: Infinity,
inputPrice: 2.5,
outputPrice: 15,
cacheReadsPrice: 0.625,
},
],
},
"gemini-2.5-pro-preview-05-06": {
maxTokens: 65_535,
contextWindow: 1_048_576,
supportsImages: true,
supportsPromptCache: true,
inputPrice: 2.5, // This is the pricing for prompts above 200k tokens.
outputPrice: 15,
cacheReadsPrice: 0.625,
cacheWritesPrice: 4.5,
tiers: [
{
contextWindow: 200_000,
inputPrice: 1.25,
outputPrice: 10,
cacheReadsPrice: 0.31,
},
{
contextWindow: Infinity,
inputPrice: 2.5,
outputPrice: 15,
cacheReadsPrice: 0.625,
},
],
},
"gemini-2.5-pro-preview-06-05": {
maxTokens: 65_535,
contextWindow: 1_048_576,
supportsImages: true,
supportsPromptCache: true,
inputPrice: 2.5, // This is the pricing for prompts above 200k tokens.
outputPrice: 15,
cacheReadsPrice: 0.625,
cacheWritesPrice: 4.5,
maxThinkingTokens: 32_768,
supportsReasoningBudget: true,
tiers: [
{
contextWindow: 200_000,
inputPrice: 1.25,
outputPrice: 10,
cacheReadsPrice: 0.31,
},
{
contextWindow: Infinity,
inputPrice: 2.5,
outputPrice: 15,
cacheReadsPrice: 0.625,
},
],
},
"gemini-2.5-pro": {
maxTokens: 64_000,
contextWindow: 1_048_576,
Expand Down
26 changes: 0 additions & 26 deletions packages/types/src/providers/vertex.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -56,32 +56,6 @@ export const vertexModels = {
inputPrice: 0.15,
outputPrice: 0.6,
},
"gemini-2.5-pro-preview-03-25": {
maxTokens: 65_535,
contextWindow: 1_048_576,
supportsImages: true,
supportsPromptCache: true,
inputPrice: 2.5,
outputPrice: 15,
},
"gemini-2.5-pro-preview-05-06": {
maxTokens: 65_535,
contextWindow: 1_048_576,
supportsImages: true,
supportsPromptCache: true,
inputPrice: 2.5,
outputPrice: 15,
},
"gemini-2.5-pro-preview-06-05": {
maxTokens: 65_535,
contextWindow: 1_048_576,
supportsImages: true,
supportsPromptCache: true,
inputPrice: 2.5,
outputPrice: 15,
maxThinkingTokens: 32_768,
supportsReasoningBudget: true,
},
"gemini-2.5-pro": {
maxTokens: 64_000,
contextWindow: 1_048_576,
Expand Down
3 changes: 1 addition & 2 deletions src/api/providers/fetchers/__tests__/openrouter.spec.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -29,7 +29,6 @@ describe("OpenRouter API", () => {

// Define models that are intentionally excluded
const excludedModels = new Set([
"google/gemini-2.5-pro-preview", // Excluded due to lag issue (#4487)
"google/gemini-2.5-flash", // OpenRouter doesn't report this as supporting prompt caching
"google/gemini-2.5-flash-lite-preview-06-17", // OpenRouter doesn't report this as supporting prompt caching
])
Expand Down Expand Up @@ -215,7 +214,7 @@ describe("OpenRouter API", () => {
describe("getOpenRouterModelEndpoints", () => {
it("fetches model endpoints and validates schema", async () => {
const { nockDone } = await nockBack("openrouter-model-endpoints.json")
const endpoints = await getOpenRouterModelEndpoints("google/gemini-2.5-pro-preview")
const endpoints = await getOpenRouterModelEndpoints("google/gemini-2.5-pro")

expect(endpoints).toEqual({
Google: {
Expand Down
17 changes: 17 additions & 0 deletions src/api/providers/gemini.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -129,9 +129,26 @@ export class GeminiHandler extends BaseProvider implements SingleCompletionHandl
}
}

/**
* Check if the model ID is a legacy preview model that should be mapped to the GA version
*/
protected isLegacyPreviewModel(modelId: string): boolean {
return [
"gemini-2.5-pro-preview-03-25",
"gemini-2.5-pro-preview-05-06",
"gemini-2.5-pro-preview-06-05"
].includes(modelId)
}

override getModel() {
const modelId = this.options.apiModelId
let id = modelId && modelId in geminiModels ? (modelId as GeminiModelId) : geminiDefaultModelId

// Handle backward compatibility for legacy preview models
if (modelId && this.isLegacyPreviewModel(modelId)) {
id = "gemini-2.5-pro" as GeminiModelId
}

const info: ModelInfo = geminiModels[id]
const params = getModelParams({ format: "gemini", modelId: id, model: info, settings: this.options })

Expand Down
9 changes: 8 additions & 1 deletion src/api/providers/openrouter.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -69,6 +69,13 @@ export class OpenRouterHandler extends BaseProvider implements SingleCompletionH
this.client = new OpenAI({ baseURL, apiKey, defaultHeaders: DEFAULT_HEADERS })
}

/**
* Check if the model ID is a legacy Gemini preview model that should be mapped to the GA version
*/
private isLegacyGeminiPreviewModel(modelId: string): boolean {
return modelId === "google/gemini-2.5-pro-preview"
}

override async *createMessage(
systemPrompt: string,
messages: Anthropic.Messages.MessageParam[],
Expand All @@ -83,7 +90,7 @@ export class OpenRouterHandler extends BaseProvider implements SingleCompletionH
// i We should generalize this using the logic in `getModelParams`, but
// this is easier for now.
if (
(modelId === "google/gemini-2.5-pro-preview" || modelId === "google/gemini-2.5-pro") &&
(this.isLegacyGeminiPreviewModel(modelId) || modelId === "google/gemini-2.5-pro") &&
typeof reasoning === "undefined"
) {
reasoning = { exclude: true }
Expand Down
7 changes: 7 additions & 0 deletions src/api/providers/vertex.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -15,6 +15,13 @@ export class VertexHandler extends GeminiHandler implements SingleCompletionHand
override getModel() {
const modelId = this.options.apiModelId
let id = modelId && modelId in vertexModels ? (modelId as VertexModelId) : vertexDefaultModelId

// Handle backward compatibility for legacy preview models
// Use the same logic as GeminiHandler
if (modelId && this.isLegacyPreviewModel(modelId)) {
id = "gemini-2.5-pro" as VertexModelId
}

const info: ModelInfo = vertexModels[id]
const params = getModelParams({ format: "gemini", modelId: id, model: info, settings: this.options })

Expand Down
Loading