Skip to content

Update AI types #6535

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
danlapid merged 3 commits into from
Apr 10, 2026
Merged

Update AI types #6535

Show file tree
Hide file tree
Changes from 2 commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
b2ba62c
feat(types): add overload-based AI.run() with gateway fallback and ge…
samscott89 Apr 9, 2026
2635e23
Update generated snapshots
samscott89 Apr 9, 2026
312fda3
Use more specific name for AsyncResponse
samscott89 Apr 9, 2026
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
69 changes: 50 additions & 19 deletions types/defines/ai.d.ts
View file
Open in desktop
Comment thread
samscott89 marked this conversation as resolved.
Original file line number Diff line number Diff line change
Expand Up @@ -5827,6 +5827,10 @@ export declare abstract class Base_Ai_Cf_Nvidia_Nemotron_3_120B_A12B {
inputs: ChatCompletionsInput;
postProcessedOutputs: ChatCompletionsOutput;
}
export declare abstract class Base_Ai_Cf_Google_Gemma_4_26B_A4B_IT {
inputs: ChatCompletionsInput;
postProcessedOutputs: ChatCompletionsOutput;
}
export interface AiModels {
"@cf/huggingface/distilbert-sst-2-int8": BaseAiTextClassification;
"@cf/stabilityai/stable-diffusion-xl-base-1.0": BaseAiTextToImage;
Expand Down Expand Up @@ -5979,7 +5983,10 @@ export type ChatCompletionsInput = XOR<
export interface InferenceUpstreamError extends Error {}
export interface AiInternalError extends Error {}
export type AiModelListType = Record<string, any>;
export declare abstract class Ai<AiModelList extends AiModelListType = AiModels> {
export type AsyncResponse = { request_id: string };
export declare abstract class Ai<
AiModelList extends AiModelListType = AiModels,
> {
aiGatewayLogId: string | null;
gateway(gatewayId: string): AiGateway;

Expand All @@ -5997,25 +6004,49 @@ export declare abstract class Ai<AiModelList extends AiModelListType = AiModels>
* @param autoragId Instance ID
*/
autorag(autoragId: string): AutoRAG;
run<Name extends keyof AiModelList, Options extends AiOptions, InputOptions extends AiModelList[Name]["inputs"]>(

// Batch request
run<Name extends keyof AiModelList>(
model: Name,
inputs: InputOptions,
options?: Options,
): Promise<
Options extends
| {
returnRawResponse: true;
}
| {
websocket: true;
}
? Response
: InputOptions extends {
stream: true;
}
? ReadableStream
: AiModelList[Name]["postProcessedOutputs"]
>;
inputs: { requests: AiModelList[Name]['inputs'][] },
options: AiOptions & { queueRequest: true }
): Promise<AsyncResponse>;

// Raw response
run<Name extends keyof AiModelList>(
model: Name,
inputs: AiModelList[Name]['inputs'],
options: AiOptions & { returnRawResponse: true }
): Promise<Response>;

// WebSocket
run<Name extends keyof AiModelList>(
model: Name,
inputs: AiModelList[Name]['inputs'],
options: AiOptions & { websocket: true }
): Promise<Response>;

// Streaming
run<Name extends keyof AiModelList>(
model: Name,
inputs: AiModelList[Name]['inputs'] & { stream: true },
options?: AiOptions
): Promise<ReadableStream>;

// Normal (default) - known model
run<Name extends keyof AiModelList>(
model: Name,
inputs: AiModelList[Name]['inputs'],
options?: AiOptions
): Promise<AiModelList[Name]['postProcessedOutputs']>;

// Unknown model (gateway fallback)
run(
model: string & {},
inputs: Record<string, unknown>,
options?: AiOptions
): Promise<Record<string, unknown>>;

models(params?: AiModelsSearchParams): Promise<AiModelsSearchObject[]>;
toMarkdown(): ToMarkdownService;
toMarkdown(
Expand Down
74 changes: 52 additions & 22 deletions types/generated-snapshot/experimental/index.d.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -11023,6 +11023,10 @@ declare abstract class Base_Ai_Cf_Nvidia_Nemotron_3_120B_A12B {
inputs: ChatCompletionsInput;
postProcessedOutputs: ChatCompletionsOutput;
}
declare abstract class Base_Ai_Cf_Google_Gemma_4_26B_A4B_IT {
inputs: ChatCompletionsInput;
postProcessedOutputs: ChatCompletionsOutput;
}
interface AiModels {
"@cf/huggingface/distilbert-sst-2-int8": BaseAiTextClassification;
"@cf/stabilityai/stable-diffusion-xl-base-1.0": BaseAiTextToImage;
Expand Down Expand Up @@ -11178,6 +11182,9 @@ type ChatCompletionsInput = XOR<
interface InferenceUpstreamError extends Error {}
interface AiInternalError extends Error {}
type AiModelListType = Record<string, any>;
type AsyncResponse = {
request_id: string;
};
declare abstract class Ai<AiModelList extends AiModelListType = AiModels> {
aiGatewayLogId: string | null;
gateway(gatewayId: string): AiGateway;
Expand All @@ -11194,29 +11201,52 @@ declare abstract class Ai<AiModelList extends AiModelListType = AiModels> {
* @param autoragId Instance ID
*/
autorag(autoragId: string): AutoRAG;
run<
Name extends keyof AiModelList,
Options extends AiOptions,
InputOptions extends AiModelList[Name]["inputs"],
>(
// Batch request
run<Name extends keyof AiModelList>(
model: Name,
inputs: InputOptions,
options?: Options,
): Promise<
Options extends
| {
returnRawResponse: true;
}
| {
websocket: true;
}
? Response
: InputOptions extends {
stream: true;
}
? ReadableStream
: AiModelList[Name]["postProcessedOutputs"]
>;
inputs: {
requests: AiModelList[Name]["inputs"][];
},
options: AiOptions & {
queueRequest: true;
},
): Promise<AsyncResponse>;
// Raw response
run<Name extends keyof AiModelList>(
model: Name,
inputs: AiModelList[Name]["inputs"],
options: AiOptions & {
returnRawResponse: true;
},
): Promise<Response>;
// WebSocket
run<Name extends keyof AiModelList>(
model: Name,
inputs: AiModelList[Name]["inputs"],
options: AiOptions & {
websocket: true;
},
): Promise<Response>;
// Streaming
run<Name extends keyof AiModelList>(
model: Name,
inputs: AiModelList[Name]["inputs"] & {
stream: true;
},
options?: AiOptions,
): Promise<ReadableStream>;
// Normal (default) - known model
run<Name extends keyof AiModelList>(
model: Name,
inputs: AiModelList[Name]["inputs"],
options?: AiOptions,
): Promise<AiModelList[Name]["postProcessedOutputs"]>;
// Unknown model (gateway fallback)
run(
model: string & {},
inputs: Record<string, unknown>,
options?: AiOptions,
): Promise<Record<string, unknown>>;
models(params?: AiModelsSearchParams): Promise<AiModelsSearchObject[]>;
toMarkdown(): ToMarkdownService;
toMarkdown(
Expand Down
74 changes: 52 additions & 22 deletions types/generated-snapshot/experimental/index.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -11032,6 +11032,10 @@ export declare abstract class Base_Ai_Cf_Nvidia_Nemotron_3_120B_A12B {
inputs: ChatCompletionsInput;
postProcessedOutputs: ChatCompletionsOutput;
}
export declare abstract class Base_Ai_Cf_Google_Gemma_4_26B_A4B_IT {
inputs: ChatCompletionsInput;
postProcessedOutputs: ChatCompletionsOutput;
}
export interface AiModels {
"@cf/huggingface/distilbert-sst-2-int8": BaseAiTextClassification;
"@cf/stabilityai/stable-diffusion-xl-base-1.0": BaseAiTextToImage;
Expand Down Expand Up @@ -11187,6 +11191,9 @@ export type ChatCompletionsInput = XOR<
export interface InferenceUpstreamError extends Error {}
export interface AiInternalError extends Error {}
export type AiModelListType = Record<string, any>;
export type AsyncResponse = {
request_id: string;
};
export declare abstract class Ai<
AiModelList extends AiModelListType = AiModels,
> {
Expand All @@ -11205,29 +11212,52 @@ export declare abstract class Ai<
* @param autoragId Instance ID
*/
autorag(autoragId: string): AutoRAG;
run<
Name extends keyof AiModelList,
Options extends AiOptions,
InputOptions extends AiModelList[Name]["inputs"],
>(
// Batch request
run<Name extends keyof AiModelList>(
model: Name,
inputs: InputOptions,
options?: Options,
): Promise<
Options extends
| {
returnRawResponse: true;
}
| {
websocket: true;
}
? Response
: InputOptions extends {
stream: true;
}
? ReadableStream
: AiModelList[Name]["postProcessedOutputs"]
>;
inputs: {
requests: AiModelList[Name]["inputs"][];
},
options: AiOptions & {
queueRequest: true;
},
): Promise<AsyncResponse>;
// Raw response
run<Name extends keyof AiModelList>(
model: Name,
inputs: AiModelList[Name]["inputs"],
options: AiOptions & {
returnRawResponse: true;
},
): Promise<Response>;
// WebSocket
run<Name extends keyof AiModelList>(
model: Name,
inputs: AiModelList[Name]["inputs"],
options: AiOptions & {
websocket: true;
},
): Promise<Response>;
// Streaming
run<Name extends keyof AiModelList>(
model: Name,
inputs: AiModelList[Name]["inputs"] & {
stream: true;
},
options?: AiOptions,
): Promise<ReadableStream>;
// Normal (default) - known model
run<Name extends keyof AiModelList>(
model: Name,
inputs: AiModelList[Name]["inputs"],
options?: AiOptions,
): Promise<AiModelList[Name]["postProcessedOutputs"]>;
// Unknown model (gateway fallback)
run(
model: string & {},
inputs: Record<string, unknown>,
options?: AiOptions,
): Promise<Record<string, unknown>>;
models(params?: AiModelsSearchParams): Promise<AiModelsSearchObject[]>;
toMarkdown(): ToMarkdownService;
toMarkdown(
Expand Down
74 changes: 52 additions & 22 deletions types/generated-snapshot/latest/index.d.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -10331,6 +10331,10 @@ declare abstract class Base_Ai_Cf_Nvidia_Nemotron_3_120B_A12B {
inputs: ChatCompletionsInput;
postProcessedOutputs: ChatCompletionsOutput;
}
declare abstract class Base_Ai_Cf_Google_Gemma_4_26B_A4B_IT {
inputs: ChatCompletionsInput;
postProcessedOutputs: ChatCompletionsOutput;
}
interface AiModels {
"@cf/huggingface/distilbert-sst-2-int8": BaseAiTextClassification;
"@cf/stabilityai/stable-diffusion-xl-base-1.0": BaseAiTextToImage;
Expand Down Expand Up @@ -10486,6 +10490,9 @@ type ChatCompletionsInput = XOR<
interface InferenceUpstreamError extends Error {}
interface AiInternalError extends Error {}
type AiModelListType = Record<string, any>;
type AsyncResponse = {
Comment thread
samscott89 marked this conversation as resolved.
Outdated
request_id: string;
};
declare abstract class Ai<AiModelList extends AiModelListType = AiModels> {
aiGatewayLogId: string | null;
gateway(gatewayId: string): AiGateway;
Expand All @@ -10502,29 +10509,52 @@ declare abstract class Ai<AiModelList extends AiModelListType = AiModels> {
* @param autoragId Instance ID
*/
autorag(autoragId: string): AutoRAG;
run<
Name extends keyof AiModelList,
Options extends AiOptions,
InputOptions extends AiModelList[Name]["inputs"],
>(
// Batch request
run<Name extends keyof AiModelList>(
model: Name,
inputs: InputOptions,
options?: Options,
): Promise<
Options extends
| {
returnRawResponse: true;
}
| {
websocket: true;
}
? Response
: InputOptions extends {
stream: true;
}
? ReadableStream
: AiModelList[Name]["postProcessedOutputs"]
>;
inputs: {
requests: AiModelList[Name]["inputs"][];
},
options: AiOptions & {
queueRequest: true;
},
): Promise<AsyncResponse>;
// Raw response
run<Name extends keyof AiModelList>(
model: Name,
inputs: AiModelList[Name]["inputs"],
options: AiOptions & {
returnRawResponse: true;
},
): Promise<Response>;
// WebSocket
run<Name extends keyof AiModelList>(
model: Name,
inputs: AiModelList[Name]["inputs"],
options: AiOptions & {
websocket: true;
},
): Promise<Response>;
// Streaming
run<Name extends keyof AiModelList>(
model: Name,
inputs: AiModelList[Name]["inputs"] & {
stream: true;
},
options?: AiOptions,
): Promise<ReadableStream>;
// Normal (default) - known model
run<Name extends keyof AiModelList>(
model: Name,
inputs: AiModelList[Name]["inputs"],
options?: AiOptions,
): Promise<AiModelList[Name]["postProcessedOutputs"]>;
// Unknown model (gateway fallback)
run(
model: string & {},
inputs: Record<string, unknown>,
options?: AiOptions,
): Promise<Record<string, unknown>>;
models(params?: AiModelsSearchParams): Promise<AiModelsSearchObject[]>;
toMarkdown(): ToMarkdownService;
toMarkdown(
Expand Down
Loading
Loading