cloudflare 6.0.0-beta.1 → 6.0.0

package/src/resources/aisearch/instances/instances.ts

@@ -2,15 +2,6 @@
 
 import { APIResource } from '../../../resource';
 import * as Core from '../../../core';
-import * as ItemsAPI from './items';
-import {
-  ItemGetParams,
-  ItemGetResponse,
-  ItemListParams,
-  ItemListResponse,
-  ItemListResponsesV4PagePaginationArray,
-  Items,
-} from './items';
 import * as JobsAPI from './jobs';
 import {
   JobCreateParams,
@@ -28,20 +19,16 @@ import * as SippyAPI from '../../r2/buckets/sippy';
 import { V4PagePaginationArray, type V4PagePaginationArrayParams } from '../../../pagination';
 
 export class Instances extends APIResource {
-  items: ItemsAPI.Items = new ItemsAPI.Items(this._client);
   jobs: JobsAPI.Jobs = new JobsAPI.Jobs(this._client);
 
   /**
-   * Create new instances.
+   * Create a new instance.
    *
    * @example
    * ```ts
    * const instance = await client.aiSearch.instances.create({
    *   account_id: 'c3dc5f0b34a14ff8e1b3ec04895e1b22',
    *   id: 'my-ai-search',
-   *   source: 'source',
-   *   token_id: '182bd5e5-6e1a-4fe4-a799-aa6d9a6ab26e',
-   *   type: 'r2',
    * });
    * ```
    */
@@ -59,7 +46,7 @@ export class Instances extends APIResource {
   }
 
   /**
-   * Update instances.
+   * Update instance.
    *
    * @example
    * ```ts
@@ -109,7 +96,7 @@ export class Instances extends APIResource {
   }
 
   /**
-   * Delete instances.
+   * Delete instance.
    *
    * @example
    * ```ts
@@ -133,7 +120,35 @@ export class Instances extends APIResource {
   }
 
   /**
-   * Read instances.
+   * Performs a chat completion request against an AI Search instance, using indexed
+   * content as context for generating responses.
+   *
+   * @example
+   * ```ts
+   * const response =
+   *   await client.aiSearch.instances.chatCompletions(
+   *     'my-ai-search',
+   *     {
+   *       account_id: 'c3dc5f0b34a14ff8e1b3ec04895e1b22',
+   *       messages: [{ content: 'content', role: 'system' }],
+   *     },
+   *   );
+   * ```
+   */
+  chatCompletions(
+    id: string,
+    params: InstanceChatCompletionsParams,
+    options?: Core.RequestOptions,
+  ): Core.APIPromise<InstanceChatCompletionsResponse> {
+    const { account_id, ...body } = params;
+    return this._client.post(`/accounts/${account_id}/ai-search/instances/${id}/chat/completions`, {
+      body,
+      ...options,
+    });
+  }
+
+  /**
+   * Read instance.
    *
    * @example
    * ```ts
@@ -157,7 +172,33 @@ export class Instances extends APIResource {
   }
 
   /**
-   * Stats
+   * Executes a semantic search query against an AI Search instance to find relevant
+   * indexed content.
+   *
+   * @example
+   * ```ts
+   * const response = await client.aiSearch.instances.search(
+   *   'my-ai-search',
+   *   { account_id: 'c3dc5f0b34a14ff8e1b3ec04895e1b22' },
+   * );
+   * ```
+   */
+  search(
+    id: string,
+    params: InstanceSearchParams,
+    options?: Core.RequestOptions,
+  ): Core.APIPromise<InstanceSearchResponse> {
+    const { account_id, ...body } = params;
+    return (
+      this._client.post(`/accounts/${account_id}/ai-search/instances/${id}/search`, {
+        body,
+        ...options,
+      }) as Core.APIPromise<{ result: InstanceSearchResponse }>
+    )._thenUnwrap((obj) => obj.result);
+  }
+
+  /**
+   * Retrieves usage statistics for AI Search instances.
    *
    * @example
    * ```ts
@@ -186,38 +227,28 @@ export class InstanceListResponsesV4PagePaginationArray extends V4PagePagination
 
 export interface InstanceCreateResponse {
   /**
-   * Use your AI Search ID.
+   * AI Search instance ID. Lowercase alphanumeric, hyphens, and underscores.
    */
   id: string;
 
-  account_id: string;
-
-  account_tag: string;
-
   created_at: string;
 
-  internal_id: string;
-
   modified_at: string;
 
-  source: string;
-
-  token_id: string;
-
-  type: 'r2' | 'web-crawler';
-
-  vectorize_name: string;
-
-  ai_gateway_id?: string;
+  ai_gateway_id?: string | null;
 
   ai_search_model?:
     | '@cf/meta/llama-3.3-70b-instruct-fp8-fast'
+    | '@cf/zai-org/glm-4.7-flash'
     | '@cf/meta/llama-3.1-8b-instruct-fast'
     | '@cf/meta/llama-3.1-8b-instruct-fp8'
     | '@cf/meta/llama-4-scout-17b-16e-instruct'
     | '@cf/qwen/qwen3-30b-a3b-fp8'
     | '@cf/deepseek-ai/deepseek-r1-distill-qwen-32b'
     | '@cf/moonshotai/kimi-k2-instruct'
+    | '@cf/google/gemma-3-12b-it'
+    | '@cf/google/gemma-4-26b-a4b-it'
+    | '@cf/moonshotai/kimi-k2.5'
     | 'anthropic/claude-3-7-sonnet'
     | 'anthropic/claude-sonnet-4'
     | 'anthropic/claude-opus-4'
@@ -236,62 +267,86 @@ export interface InstanceCreateResponse {
     | 'openai/gpt-5'
     | 'openai/gpt-5-mini'
     | 'openai/gpt-5-nano'
-    | '';
+    | ''
+    | null;
 
   cache?: boolean;
 
   cache_threshold?: 'super_strict_match' | 'close_enough' | 'flexible_friend' | 'anything_goes';
 
-  chunk?: boolean;
-
   chunk_overlap?: number;
 
   chunk_size?: number;
 
-  created_by?: string;
+  created_by?: string | null;
+
+  custom_metadata?: Array<InstanceCreateResponse.CustomMetadata>;
 
   embedding_model?:
+    | '@cf/qwen/qwen3-embedding-0.6b'
     | '@cf/baai/bge-m3'
     | '@cf/baai/bge-large-en-v1.5'
     | '@cf/google/embeddinggemma-300m'
-    | '@cf/qwen/qwen3-embedding-0.6b'
     | 'google-ai-studio/gemini-embedding-001'
+    | 'google-ai-studio/gemini-embedding-2-preview'
     | 'openai/text-embedding-3-small'
     | 'openai/text-embedding-3-large'
-    | '';
+    | ''
+    | null;
 
   enable?: boolean;
 
   engine_version?: number;
 
+  fusion_method?: 'max' | 'rrf';
+
+  /**
+   * @deprecated Deprecated — use index_method instead.
+   */
   hybrid_search_enabled?: boolean;
 
-  last_activity?: string;
+  /**
+   * Controls which storage backends are used during indexing. Defaults to
+   * vector-only.
+   */
+  index_method?: InstanceCreateResponse.IndexMethod;
+
+  indexing_options?: InstanceCreateResponse.IndexingOptions | null;
+
+  last_activity?: string | null;
 
   max_num_results?: number;
 
   metadata?: InstanceCreateResponse.Metadata;
 
-  modified_by?: string;
+  modified_by?: string | null;
+
+  namespace?: string | null;
 
   paused?: boolean;
 
-  public_endpoint_id?: string;
+  public_endpoint_id?: string | null;
 
   public_endpoint_params?: InstanceCreateResponse.PublicEndpointParams;
 
   reranking?: boolean;
 
-  reranking_model?: '@cf/baai/bge-reranker-base' | '';
+  reranking_model?: '@cf/baai/bge-reranker-base' | '' | null;
+
+  retrieval_options?: InstanceCreateResponse.RetrievalOptions | null;
 
   rewrite_model?:
     | '@cf/meta/llama-3.3-70b-instruct-fp8-fast'
+    | '@cf/zai-org/glm-4.7-flash'
     | '@cf/meta/llama-3.1-8b-instruct-fast'
     | '@cf/meta/llama-3.1-8b-instruct-fp8'
     | '@cf/meta/llama-4-scout-17b-16e-instruct'
     | '@cf/qwen/qwen3-30b-a3b-fp8'
     | '@cf/deepseek-ai/deepseek-r1-distill-qwen-32b'
     | '@cf/moonshotai/kimi-k2-instruct'
+    | '@cf/google/gemma-3-12b-it'
+    | '@cf/google/gemma-4-26b-a4b-it'
+    | '@cf/moonshotai/kimi-k2.5'
     | 'anthropic/claude-3-7-sonnet'
     | 'anthropic/claude-sonnet-4'
     | 'anthropic/claude-opus-4'
@@ -310,56 +365,63 @@ export interface InstanceCreateResponse {
     | 'openai/gpt-5'
     | 'openai/gpt-5-mini'
     | 'openai/gpt-5-nano'
-    | '';
+    | ''
+    | null;
 
   rewrite_query?: boolean;
 
   score_threshold?: number;
 
-  source_params?: InstanceCreateResponse.SourceParams;
+  source?: string | null;
+
+  source_params?: InstanceCreateResponse.SourceParams | null;
 
   status?: string;
 
-  summarization?: boolean;
+  /**
+   * Interval between automatic syncs, in seconds. Allowed values: 900 (15min), 1800
+   * (30min), 3600 (1h), 7200 (2h), 14400 (4h), 21600 (6h), 43200 (12h), 86400 (24h).
+   */
+  sync_interval?: 900 | 1800 | 3600 | 7200 | 14400 | 21600 | 43200 | 86400;
 
-  summarization_model?:
-    | '@cf/meta/llama-3.3-70b-instruct-fp8-fast'
-    | '@cf/meta/llama-3.1-8b-instruct-fast'
-    | '@cf/meta/llama-3.1-8b-instruct-fp8'
-    | '@cf/meta/llama-4-scout-17b-16e-instruct'
-    | '@cf/qwen/qwen3-30b-a3b-fp8'
-    | '@cf/deepseek-ai/deepseek-r1-distill-qwen-32b'
-    | '@cf/moonshotai/kimi-k2-instruct'
-    | 'anthropic/claude-3-7-sonnet'
-    | 'anthropic/claude-sonnet-4'
-    | 'anthropic/claude-opus-4'
-    | 'anthropic/claude-3-5-haiku'
-    | 'cerebras/qwen-3-235b-a22b-instruct'
-    | 'cerebras/qwen-3-235b-a22b-thinking'
-    | 'cerebras/llama-3.3-70b'
-    | 'cerebras/llama-4-maverick-17b-128e-instruct'
-    | 'cerebras/llama-4-scout-17b-16e-instruct'
-    | 'cerebras/gpt-oss-120b'
-    | 'google-ai-studio/gemini-2.5-flash'
-    | 'google-ai-studio/gemini-2.5-pro'
-    | 'grok/grok-4'
-    | 'groq/llama-3.3-70b-versatile'
-    | 'groq/llama-3.1-8b-instant'
-    | 'openai/gpt-5'
-    | 'openai/gpt-5-mini'
-    | 'openai/gpt-5-nano'
-    | '';
+  token_id?: string;
 
-  system_prompt_ai_search?: string;
+  type?: 'r2' | 'web-crawler' | null;
+}
 
-  system_prompt_index_summarization?: string;
+export namespace InstanceCreateResponse {
+  export interface CustomMetadata {
+    data_type: 'text' | 'number' | 'boolean' | 'datetime';
 
-  system_prompt_rewrite_query?: string;
+    field_name: string;
+  }
 
-  vectorize_active_namespace?: string;
-}
+  /**
+   * Controls which storage backends are used during indexing. Defaults to
+   * vector-only.
+   */
+  export interface IndexMethod {
+    /**
+     * Enable keyword (BM25) storage backend.
+     */
+    keyword: boolean;
+
+    /**
+     * Enable vector (embedding) storage backend.
+     */
+    vector: boolean;
+  }
+
+  export interface IndexingOptions {
+    /**
+     * Tokenizer used for keyword search indexing. porter provides word-level
+     * tokenization with Porter stemming (good for natural language queries). trigram
+     * enables character-level substring matching (good for partial matches, code,
+     * identifiers). Changing this triggers a full re-index. Defaults to porter.
+     */
+    keyword_tokenizer?: 'porter' | 'trigram';
+  }
 
-export namespace InstanceCreateResponse {
   export interface Metadata {
     created_from_aisearch_wizard?: boolean;
 
@@ -389,6 +451,8 @@ export namespace InstanceCreateResponse {
     }
 
     export interface Mcp {
+      description?: string;
+
       /**
        * Disable MCP endpoint for this public endpoint
        */
@@ -411,16 +475,54 @@ export namespace InstanceCreateResponse {
     }
   }
 
+  export interface RetrievalOptions {
+    /**
+     * Metadata fields to boost search results by. Each entry specifies a metadata
+     * field and an optional direction. Direction defaults to 'asc' for numeric fields
+     * and 'exists' for text/boolean fields. Fields must match 'timestamp' or a defined
+     * custom_metadata field.
+     */
+    boost_by?: Array<RetrievalOptions.BoostBy>;
+
+    /**
+     * Controls which documents are candidates for BM25 scoring. 'and' restricts
+     * candidates to documents containing all query terms; 'or' includes any document
+     * containing at least one term, ranked by BM25 relevance. Defaults to 'and'.
+     */
+    keyword_match_mode?: 'and' | 'or';
+  }
+
+  export namespace RetrievalOptions {
+    export interface BoostBy {
+      /**
+       * Metadata field name to boost by. Use 'timestamp' for document freshness, or any
+       * custom_metadata field. Numeric and datetime fields support asc/desc directions;
+       * text/boolean fields support exists/not_exists.
+       */
+      field: string;
+
+      /**
+       * Boost direction. 'desc' = higher values rank higher (e.g. newer timestamps).
+       * 'asc' = lower values rank higher. 'exists' = boost chunks that have the field.
+       * 'not_exists' = boost chunks that lack the field. Optional ��� defaults to 'asc'
+       * for numeric/datetime fields, 'exists' for text/boolean fields.
+       */
+      direction?: 'asc' | 'desc' | 'exists' | 'not_exists';
+    }
+  }
+
   export interface SourceParams {
     /**
-     * List of path patterns to exclude. Supports wildcards (e.g., _/admin/_,
-     * /private/\*_, _\private\*)
+     * List of path patterns to exclude. Uses micromatch glob syntax: \* matches within
+     * a path segment, ** matches across path segments (e.g., /admin/** matches
+     * /admin/users and /admin/settings/advanced)
      */
     exclude_items?: Array<string>;
 
     /**
-     * List of path patterns to include. Supports wildcards (e.g., _/blog/_.html,
-     * /docs/\*_, _\blog\*.html)
+     * List of path patterns to include. Uses micromatch glob syntax: \* matches within
+     * a path segment, ** matches across path segments (e.g., /blog/** matches
+     * /blog/post and /blog/2024/post)
      */
     include_items?: Array<string>;
 
@@ -433,22 +535,65 @@ export namespace InstanceCreateResponse {
 
   export namespace SourceParams {
     export interface WebCrawler {
+      crawl_options?: WebCrawler.CrawlOptions;
+
       parse_options?: WebCrawler.ParseOptions;
 
-      parse_type?: 'sitemap' | 'feed-rss';
+      parse_type?: 'sitemap' | 'feed-rss' | 'crawl';
 
       store_options?: WebCrawler.StoreOptions;
     }
 
     export namespace WebCrawler {
+      export interface CrawlOptions {
+        depth?: number;
+
+        include_external_links?: boolean;
+
+        include_subdomains?: boolean;
+
+        max_age?: number;
+
+        source?: 'all' | 'sitemaps' | 'links';
+      }
+
       export interface ParseOptions {
+        /**
+         * List of path-to-selector mappings for extracting specific content from crawled
+         * pages. Each entry pairs a URL glob pattern with a CSS selector. The first
+         * matching path wins. Only the matched HTML fragment is stored and indexed.
+         */
+        content_selector?: Array<ParseOptions.ContentSelector>;
+
         include_headers?: { [key: string]: string };
 
         include_images?: boolean;
 
+        /**
+         * List of specific sitemap URLs to use for crawling. Only valid when parse_type is
+         * 'sitemap'.
+         */
+        specific_sitemaps?: Array<string>;
+
         use_browser_rendering?: boolean;
       }
 
+      export namespace ParseOptions {
+        export interface ContentSelector {
+          /**
+           * Glob pattern to match against the page URL path. Uses standard glob syntax: \*
+           * matches within a segment, \*\* crosses directories.
+           */
+          path: string;
+
+          /**
+           * CSS selector to extract content from pages matching the path pattern. Supports
+           * standard CSS selectors including class, ID, element, and attribute selectors.
+           */
+          selector: string;
+        }
+      }
+
       export interface StoreOptions {
         storage_id: string;
 
@@ -462,38 +607,28 @@ export namespace InstanceCreateResponse {
 
 export interface InstanceUpdateResponse {
   /**
-   * Use your AI Search ID.
+   * AI Search instance ID. Lowercase alphanumeric, hyphens, and underscores.
    */
   id: string;
 
-  account_id: string;
-
-  account_tag: string;
-
   created_at: string;
 
-  internal_id: string;
-
   modified_at: string;
 
-  source: string;
-
-  token_id: string;
-
-  type: 'r2' | 'web-crawler';
-
-  vectorize_name: string;
-
-  ai_gateway_id?: string;
+  ai_gateway_id?: string | null;
 
   ai_search_model?:
     | '@cf/meta/llama-3.3-70b-instruct-fp8-fast'
+    | '@cf/zai-org/glm-4.7-flash'
     | '@cf/meta/llama-3.1-8b-instruct-fast'
     | '@cf/meta/llama-3.1-8b-instruct-fp8'
     | '@cf/meta/llama-4-scout-17b-16e-instruct'
     | '@cf/qwen/qwen3-30b-a3b-fp8'
     | '@cf/deepseek-ai/deepseek-r1-distill-qwen-32b'
     | '@cf/moonshotai/kimi-k2-instruct'
+    | '@cf/google/gemma-3-12b-it'
+    | '@cf/google/gemma-4-26b-a4b-it'
+    | '@cf/moonshotai/kimi-k2.5'
     | 'anthropic/claude-3-7-sonnet'
     | 'anthropic/claude-sonnet-4'
     | 'anthropic/claude-opus-4'
@@ -512,62 +647,86 @@ export interface InstanceUpdateResponse {
     | 'openai/gpt-5'
     | 'openai/gpt-5-mini'
     | 'openai/gpt-5-nano'
-    | '';
+    | ''
+    | null;
 
   cache?: boolean;
 
   cache_threshold?: 'super_strict_match' | 'close_enough' | 'flexible_friend' | 'anything_goes';
 
-  chunk?: boolean;
-
   chunk_overlap?: number;
 
   chunk_size?: number;
 
-  created_by?: string;
+  created_by?: string | null;
+
+  custom_metadata?: Array<InstanceUpdateResponse.CustomMetadata>;
 
   embedding_model?:
+    | '@cf/qwen/qwen3-embedding-0.6b'
     | '@cf/baai/bge-m3'
     | '@cf/baai/bge-large-en-v1.5'
     | '@cf/google/embeddinggemma-300m'
-    | '@cf/qwen/qwen3-embedding-0.6b'
     | 'google-ai-studio/gemini-embedding-001'
+    | 'google-ai-studio/gemini-embedding-2-preview'
     | 'openai/text-embedding-3-small'
     | 'openai/text-embedding-3-large'
-    | '';
+    | ''
+    | null;
 
   enable?: boolean;
 
   engine_version?: number;
 
+  fusion_method?: 'max' | 'rrf';
+
+  /**
+   * @deprecated Deprecated — use index_method instead.
+   */
   hybrid_search_enabled?: boolean;
 
-  last_activity?: string;
+  /**
+   * Controls which storage backends are used during indexing. Defaults to
+   * vector-only.
+   */
+  index_method?: InstanceUpdateResponse.IndexMethod;
+
+  indexing_options?: InstanceUpdateResponse.IndexingOptions | null;
+
+  last_activity?: string | null;
 
   max_num_results?: number;
 
   metadata?: InstanceUpdateResponse.Metadata;
 
-  modified_by?: string;
+  modified_by?: string | null;
+
+  namespace?: string | null;
 
   paused?: boolean;
 
-  public_endpoint_id?: string;
+  public_endpoint_id?: string | null;
 
   public_endpoint_params?: InstanceUpdateResponse.PublicEndpointParams;
 
   reranking?: boolean;
 
-  reranking_model?: '@cf/baai/bge-reranker-base' | '';
+  reranking_model?: '@cf/baai/bge-reranker-base' | '' | null;
+
+  retrieval_options?: InstanceUpdateResponse.RetrievalOptions | null;
 
   rewrite_model?:
     | '@cf/meta/llama-3.3-70b-instruct-fp8-fast'
+    | '@cf/zai-org/glm-4.7-flash'
     | '@cf/meta/llama-3.1-8b-instruct-fast'
     | '@cf/meta/llama-3.1-8b-instruct-fp8'
     | '@cf/meta/llama-4-scout-17b-16e-instruct'
     | '@cf/qwen/qwen3-30b-a3b-fp8'
     | '@cf/deepseek-ai/deepseek-r1-distill-qwen-32b'
     | '@cf/moonshotai/kimi-k2-instruct'
+    | '@cf/google/gemma-3-12b-it'
+    | '@cf/google/gemma-4-26b-a4b-it'
+    | '@cf/moonshotai/kimi-k2.5'
     | 'anthropic/claude-3-7-sonnet'
     | 'anthropic/claude-sonnet-4'
     | 'anthropic/claude-opus-4'
@@ -586,56 +745,63 @@ export interface InstanceUpdateResponse {
     | 'openai/gpt-5'
     | 'openai/gpt-5-mini'
     | 'openai/gpt-5-nano'
-    | '';
+    | ''
+    | null;
 
   rewrite_query?: boolean;
 
   score_threshold?: number;
 
-  source_params?: InstanceUpdateResponse.SourceParams;
+  source?: string | null;
+
+  source_params?: InstanceUpdateResponse.SourceParams | null;
 
   status?: string;
 
-  summarization?: boolean;
+  /**
+   * Interval between automatic syncs, in seconds. Allowed values: 900 (15min), 1800
+   * (30min), 3600 (1h), 7200 (2h), 14400 (4h), 21600 (6h), 43200 (12h), 86400 (24h).
+   */
+  sync_interval?: 900 | 1800 | 3600 | 7200 | 14400 | 21600 | 43200 | 86400;
 
-  summarization_model?:
-    | '@cf/meta/llama-3.3-70b-instruct-fp8-fast'
-    | '@cf/meta/llama-3.1-8b-instruct-fast'
-    | '@cf/meta/llama-3.1-8b-instruct-fp8'
-    | '@cf/meta/llama-4-scout-17b-16e-instruct'
-    | '@cf/qwen/qwen3-30b-a3b-fp8'
-    | '@cf/deepseek-ai/deepseek-r1-distill-qwen-32b'
-    | '@cf/moonshotai/kimi-k2-instruct'
-    | 'anthropic/claude-3-7-sonnet'
-    | 'anthropic/claude-sonnet-4'
-    | 'anthropic/claude-opus-4'
-    | 'anthropic/claude-3-5-haiku'
-    | 'cerebras/qwen-3-235b-a22b-instruct'
-    | 'cerebras/qwen-3-235b-a22b-thinking'
-    | 'cerebras/llama-3.3-70b'
-    | 'cerebras/llama-4-maverick-17b-128e-instruct'
-    | 'cerebras/llama-4-scout-17b-16e-instruct'
-    | 'cerebras/gpt-oss-120b'
-    | 'google-ai-studio/gemini-2.5-flash'
-    | 'google-ai-studio/gemini-2.5-pro'
-    | 'grok/grok-4'
-    | 'groq/llama-3.3-70b-versatile'
-    | 'groq/llama-3.1-8b-instant'
-    | 'openai/gpt-5'
-    | 'openai/gpt-5-mini'
-    | 'openai/gpt-5-nano'
-    | '';
+  token_id?: string;
+
+  type?: 'r2' | 'web-crawler' | null;
+}
 
-  system_prompt_ai_search?: string;
+export namespace InstanceUpdateResponse {
+  export interface CustomMetadata {
+    data_type: 'text' | 'number' | 'boolean' | 'datetime';
 
-  system_prompt_index_summarization?: string;
+    field_name: string;
+  }
 
-  system_prompt_rewrite_query?: string;
+  /**
+   * Controls which storage backends are used during indexing. Defaults to
+   * vector-only.
+   */
+  export interface IndexMethod {
+    /**
+     * Enable keyword (BM25) storage backend.
+     */
+    keyword: boolean;
 
-  vectorize_active_namespace?: string;
-}
+    /**
+     * Enable vector (embedding) storage backend.
+     */
+    vector: boolean;
+  }
+
+  export interface IndexingOptions {
+    /**
+     * Tokenizer used for keyword search indexing. porter provides word-level
+     * tokenization with Porter stemming (good for natural language queries). trigram
+     * enables character-level substring matching (good for partial matches, code,
+     * identifiers). Changing this triggers a full re-index. Defaults to porter.
+     */
+    keyword_tokenizer?: 'porter' | 'trigram';
+  }
 
-export namespace InstanceUpdateResponse {
   export interface Metadata {
     created_from_aisearch_wizard?: boolean;
 
@@ -665,6 +831,8 @@ export namespace InstanceUpdateResponse {
     }
 
     export interface Mcp {
+      description?: string;
+
       /**
        * Disable MCP endpoint for this public endpoint
        */
@@ -687,16 +855,54 @@ export namespace InstanceUpdateResponse {
     }
   }
 
+  export interface RetrievalOptions {
+    /**
+     * Metadata fields to boost search results by. Each entry specifies a metadata
+     * field and an optional direction. Direction defaults to 'asc' for numeric fields
+     * and 'exists' for text/boolean fields. Fields must match 'timestamp' or a defined
+     * custom_metadata field.
+     */
+    boost_by?: Array<RetrievalOptions.BoostBy>;
+
+    /**
+     * Controls which documents are candidates for BM25 scoring. 'and' restricts
+     * candidates to documents containing all query terms; 'or' includes any document
+     * containing at least one term, ranked by BM25 relevance. Defaults to 'and'.
+     */
+    keyword_match_mode?: 'and' | 'or';
+  }
+
+  export namespace RetrievalOptions {
+    export interface BoostBy {
+      /**
+       * Metadata field name to boost by. Use 'timestamp' for document freshness, or any
+       * custom_metadata field. Numeric and datetime fields support asc/desc directions;
+       * text/boolean fields support exists/not_exists.
+       */
+      field: string;
+
+      /**
+       * Boost direction. 'desc' = higher values rank higher (e.g. newer timestamps).
+       * 'asc' = lower values rank higher. 'exists' = boost chunks that have the field.
+       * 'not_exists' = boost chunks that lack the field. Optional ��� defaults to 'asc'
+       * for numeric/datetime fields, 'exists' for text/boolean fields.
+       */
+      direction?: 'asc' | 'desc' | 'exists' | 'not_exists';
+    }
+  }
+
   export interface SourceParams {
     /**
-     * List of path patterns to exclude. Supports wildcards (e.g., _/admin/_,
-     * /private/\*_, _\private\*)
+     * List of path patterns to exclude. Uses micromatch glob syntax: \* matches within
+     * a path segment, ** matches across path segments (e.g., /admin/** matches
+     * /admin/users and /admin/settings/advanced)
      */
     exclude_items?: Array<string>;
 
     /**
-     * List of path patterns to include. Supports wildcards (e.g., _/blog/_.html,
-     * /docs/\*_, _\blog\*.html)
+     * List of path patterns to include. Uses micromatch glob syntax: \* matches within
+     * a path segment, ** matches across path segments (e.g., /blog/** matches
+     * /blog/post and /blog/2024/post)
      */
     include_items?: Array<string>;
 
@@ -709,22 +915,65 @@ export namespace InstanceUpdateResponse {
 
   export namespace SourceParams {
     export interface WebCrawler {
+      crawl_options?: WebCrawler.CrawlOptions;
+
       parse_options?: WebCrawler.ParseOptions;
 
-      parse_type?: 'sitemap' | 'feed-rss';
+      parse_type?: 'sitemap' | 'feed-rss' | 'crawl';
 
       store_options?: WebCrawler.StoreOptions;
     }
 
     export namespace WebCrawler {
+      export interface CrawlOptions {
+        depth?: number;
+
+        include_external_links?: boolean;
+
+        include_subdomains?: boolean;
+
+        max_age?: number;
+
+        source?: 'all' | 'sitemaps' | 'links';
+      }
+
       export interface ParseOptions {
+        /**
+         * List of path-to-selector mappings for extracting specific content from crawled
+         * pages. Each entry pairs a URL glob pattern with a CSS selector. The first
+         * matching path wins. Only the matched HTML fragment is stored and indexed.
+         */
+        content_selector?: Array<ParseOptions.ContentSelector>;
+
         include_headers?: { [key: string]: string };
 
         include_images?: boolean;
 
+        /**
+         * List of specific sitemap URLs to use for crawling. Only valid when parse_type is
+         * 'sitemap'.
+         */
+        specific_sitemaps?: Array<string>;
+
         use_browser_rendering?: boolean;
       }
 
+      export namespace ParseOptions {
+        export interface ContentSelector {
+          /**
+           * Glob pattern to match against the page URL path. Uses standard glob syntax: \*
+           * matches within a segment, \*\* crosses directories.
+           */
+          path: string;
+
+          /**
+           * CSS selector to extract content from pages matching the path pattern. Supports
+           * standard CSS selectors including class, ID, element, and attribute selectors.
+           */
+          selector: string;
+        }
+      }
+
       export interface StoreOptions {
         storage_id: string;
 
@@ -738,38 +987,28 @@ export namespace InstanceUpdateResponse {
 
 export interface InstanceListResponse {
   /**
-   * Use your AI Search ID.
+   * AI Search instance ID. Lowercase alphanumeric, hyphens, and underscores.
    */
   id: string;
 
-  account_id: string;
-
-  account_tag: string;
-
   created_at: string;
 
-  internal_id: string;
-
   modified_at: string;
 
-  source: string;
-
-  token_id: string;
-
-  type: 'r2' | 'web-crawler';
-
-  vectorize_name: string;
-
-  ai_gateway_id?: string;
+  ai_gateway_id?: string | null;
 
   ai_search_model?:
     | '@cf/meta/llama-3.3-70b-instruct-fp8-fast'
+    | '@cf/zai-org/glm-4.7-flash'
     | '@cf/meta/llama-3.1-8b-instruct-fast'
     | '@cf/meta/llama-3.1-8b-instruct-fp8'
     | '@cf/meta/llama-4-scout-17b-16e-instruct'
     | '@cf/qwen/qwen3-30b-a3b-fp8'
     | '@cf/deepseek-ai/deepseek-r1-distill-qwen-32b'
     | '@cf/moonshotai/kimi-k2-instruct'
+    | '@cf/google/gemma-3-12b-it'
+    | '@cf/google/gemma-4-26b-a4b-it'
+    | '@cf/moonshotai/kimi-k2.5'
     | 'anthropic/claude-3-7-sonnet'
     | 'anthropic/claude-sonnet-4'
     | 'anthropic/claude-opus-4'
@@ -788,62 +1027,86 @@ export interface InstanceListResponse {
     | 'openai/gpt-5'
     | 'openai/gpt-5-mini'
     | 'openai/gpt-5-nano'
-    | '';
+    | ''
+    | null;
 
   cache?: boolean;
 
   cache_threshold?: 'super_strict_match' | 'close_enough' | 'flexible_friend' | 'anything_goes';
 
-  chunk?: boolean;
-
   chunk_overlap?: number;
 
   chunk_size?: number;
 
-  created_by?: string;
+  created_by?: string | null;
+
+  custom_metadata?: Array<InstanceListResponse.CustomMetadata>;
 
   embedding_model?:
+    | '@cf/qwen/qwen3-embedding-0.6b'
     | '@cf/baai/bge-m3'
     | '@cf/baai/bge-large-en-v1.5'
     | '@cf/google/embeddinggemma-300m'
-    | '@cf/qwen/qwen3-embedding-0.6b'
     | 'google-ai-studio/gemini-embedding-001'
+    | 'google-ai-studio/gemini-embedding-2-preview'
     | 'openai/text-embedding-3-small'
     | 'openai/text-embedding-3-large'
-    | '';
+    | ''
+    | null;
 
   enable?: boolean;
 
   engine_version?: number;
 
+  fusion_method?: 'max' | 'rrf';
+
+  /**
+   * @deprecated Deprecated — use index_method instead.
+   */
   hybrid_search_enabled?: boolean;
 
-  last_activity?: string;
+  /**
+   * Controls which storage backends are used during indexing. Defaults to
+   * vector-only.
+   */
+  index_method?: InstanceListResponse.IndexMethod;
 
-  max_num_results?: number;
+  indexing_options?: InstanceListResponse.IndexingOptions | null;
+
+  last_activity?: string | null;
+
+  max_num_results?: number;
 
   metadata?: InstanceListResponse.Metadata;
 
-  modified_by?: string;
+  modified_by?: string | null;
+
+  namespace?: string | null;
 
   paused?: boolean;
 
-  public_endpoint_id?: string;
+  public_endpoint_id?: string | null;
 
   public_endpoint_params?: InstanceListResponse.PublicEndpointParams;
 
   reranking?: boolean;
 
-  reranking_model?: '@cf/baai/bge-reranker-base' | '';
+  reranking_model?: '@cf/baai/bge-reranker-base' | '' | null;
+
+  retrieval_options?: InstanceListResponse.RetrievalOptions | null;
 
   rewrite_model?:
     | '@cf/meta/llama-3.3-70b-instruct-fp8-fast'
+    | '@cf/zai-org/glm-4.7-flash'
     | '@cf/meta/llama-3.1-8b-instruct-fast'
     | '@cf/meta/llama-3.1-8b-instruct-fp8'
     | '@cf/meta/llama-4-scout-17b-16e-instruct'
     | '@cf/qwen/qwen3-30b-a3b-fp8'
     | '@cf/deepseek-ai/deepseek-r1-distill-qwen-32b'
     | '@cf/moonshotai/kimi-k2-instruct'
+    | '@cf/google/gemma-3-12b-it'
+    | '@cf/google/gemma-4-26b-a4b-it'
+    | '@cf/moonshotai/kimi-k2.5'
     | 'anthropic/claude-3-7-sonnet'
     | 'anthropic/claude-sonnet-4'
     | 'anthropic/claude-opus-4'
@@ -862,56 +1125,63 @@ export interface InstanceListResponse {
     | 'openai/gpt-5'
     | 'openai/gpt-5-mini'
     | 'openai/gpt-5-nano'
-    | '';
+    | ''
+    | null;
 
   rewrite_query?: boolean;
 
   score_threshold?: number;
 
-  source_params?: InstanceListResponse.SourceParams;
+  source?: string | null;
+
+  source_params?: InstanceListResponse.SourceParams | null;
 
   status?: string;
 
-  summarization?: boolean;
+  /**
+   * Interval between automatic syncs, in seconds. Allowed values: 900 (15min), 1800
+   * (30min), 3600 (1h), 7200 (2h), 14400 (4h), 21600 (6h), 43200 (12h), 86400 (24h).
+   */
+  sync_interval?: 900 | 1800 | 3600 | 7200 | 14400 | 21600 | 43200 | 86400;
 
-  summarization_model?:
-    | '@cf/meta/llama-3.3-70b-instruct-fp8-fast'
-    | '@cf/meta/llama-3.1-8b-instruct-fast'
-    | '@cf/meta/llama-3.1-8b-instruct-fp8'
-    | '@cf/meta/llama-4-scout-17b-16e-instruct'
-    | '@cf/qwen/qwen3-30b-a3b-fp8'
-    | '@cf/deepseek-ai/deepseek-r1-distill-qwen-32b'
-    | '@cf/moonshotai/kimi-k2-instruct'
-    | 'anthropic/claude-3-7-sonnet'
-    | 'anthropic/claude-sonnet-4'
-    | 'anthropic/claude-opus-4'
-    | 'anthropic/claude-3-5-haiku'
-    | 'cerebras/qwen-3-235b-a22b-instruct'
-    | 'cerebras/qwen-3-235b-a22b-thinking'
-    | 'cerebras/llama-3.3-70b'
-    | 'cerebras/llama-4-maverick-17b-128e-instruct'
-    | 'cerebras/llama-4-scout-17b-16e-instruct'
-    | 'cerebras/gpt-oss-120b'
-    | 'google-ai-studio/gemini-2.5-flash'
-    | 'google-ai-studio/gemini-2.5-pro'
-    | 'grok/grok-4'
-    | 'groq/llama-3.3-70b-versatile'
-    | 'groq/llama-3.1-8b-instant'
-    | 'openai/gpt-5'
-    | 'openai/gpt-5-mini'
-    | 'openai/gpt-5-nano'
-    | '';
+  token_id?: string;
 
-  system_prompt_ai_search?: string;
+  type?: 'r2' | 'web-crawler' | null;
+}
 
-  system_prompt_index_summarization?: string;
+export namespace InstanceListResponse {
+  export interface CustomMetadata {
+    data_type: 'text' | 'number' | 'boolean' | 'datetime';
 
-  system_prompt_rewrite_query?: string;
+    field_name: string;
+  }
 
-  vectorize_active_namespace?: string;
-}
+  /**
+   * Controls which storage backends are used during indexing. Defaults to
+   * vector-only.
+   */
+  export interface IndexMethod {
+    /**
+     * Enable keyword (BM25) storage backend.
+     */
+    keyword: boolean;
+
+    /**
+     * Enable vector (embedding) storage backend.
+     */
+    vector: boolean;
+  }
+
+  export interface IndexingOptions {
+    /**
+     * Tokenizer used for keyword search indexing. porter provides word-level
+     * tokenization with Porter stemming (good for natural language queries). trigram
+     * enables character-level substring matching (good for partial matches, code,
+     * identifiers). Changing this triggers a full re-index. Defaults to porter.
+     */
+    keyword_tokenizer?: 'porter' | 'trigram';
+  }
 
-export namespace InstanceListResponse {
   export interface Metadata {
     created_from_aisearch_wizard?: boolean;
 
@@ -941,6 +1211,8 @@ export namespace InstanceListResponse {
     }
 
     export interface Mcp {
+      description?: string;
+
       /**
        * Disable MCP endpoint for this public endpoint
        */
@@ -963,16 +1235,54 @@ export namespace InstanceListResponse {
     }
   }
 
+  export interface RetrievalOptions {
+    /**
+     * Metadata fields to boost search results by. Each entry specifies a metadata
+     * field and an optional direction. Direction defaults to 'asc' for numeric fields
+     * and 'exists' for text/boolean fields. Fields must match 'timestamp' or a defined
+     * custom_metadata field.
+     */
+    boost_by?: Array<RetrievalOptions.BoostBy>;
+
+    /**
+     * Controls which documents are candidates for BM25 scoring. 'and' restricts
+     * candidates to documents containing all query terms; 'or' includes any document
+     * containing at least one term, ranked by BM25 relevance. Defaults to 'and'.
+     */
+    keyword_match_mode?: 'and' | 'or';
+  }
+
+  export namespace RetrievalOptions {
+    export interface BoostBy {
+      /**
+       * Metadata field name to boost by. Use 'timestamp' for document freshness, or any
+       * custom_metadata field. Numeric and datetime fields support asc/desc directions;
+       * text/boolean fields support exists/not_exists.
+       */
+      field: string;
+
+      /**
+       * Boost direction. 'desc' = higher values rank higher (e.g. newer timestamps).
+       * 'asc' = lower values rank higher. 'exists' = boost chunks that have the field.
+       * 'not_exists' = boost chunks that lack the field. Optional ��� defaults to 'asc'
+       * for numeric/datetime fields, 'exists' for text/boolean fields.
+       */
+      direction?: 'asc' | 'desc' | 'exists' | 'not_exists';
+    }
+  }
+
   export interface SourceParams {
     /**
-     * List of path patterns to exclude. Supports wildcards (e.g., _/admin/_,
-     * /private/\*_, _\private\*)
+     * List of path patterns to exclude. Uses micromatch glob syntax: \* matches within
+     * a path segment, ** matches across path segments (e.g., /admin/** matches
+     * /admin/users and /admin/settings/advanced)
      */
     exclude_items?: Array<string>;
 
     /**
-     * List of path patterns to include. Supports wildcards (e.g., _/blog/_.html,
-     * /docs/\*_, _\blog\*.html)
+     * List of path patterns to include. Uses micromatch glob syntax: \* matches within
+     * a path segment, ** matches across path segments (e.g., /blog/** matches
+     * /blog/post and /blog/2024/post)
      */
     include_items?: Array<string>;
 
@@ -985,22 +1295,65 @@ export namespace InstanceListResponse {
 
   export namespace SourceParams {
     export interface WebCrawler {
+      crawl_options?: WebCrawler.CrawlOptions;
+
       parse_options?: WebCrawler.ParseOptions;
 
-      parse_type?: 'sitemap' | 'feed-rss';
+      parse_type?: 'sitemap' | 'feed-rss' | 'crawl';
 
       store_options?: WebCrawler.StoreOptions;
     }
 
     export namespace WebCrawler {
+      export interface CrawlOptions {
+        depth?: number;
+
+        include_external_links?: boolean;
+
+        include_subdomains?: boolean;
+
+        max_age?: number;
+
+        source?: 'all' | 'sitemaps' | 'links';
+      }
+
       export interface ParseOptions {
+        /**
+         * List of path-to-selector mappings for extracting specific content from crawled
+         * pages. Each entry pairs a URL glob pattern with a CSS selector. The first
+         * matching path wins. Only the matched HTML fragment is stored and indexed.
+         */
+        content_selector?: Array<ParseOptions.ContentSelector>;
+
         include_headers?: { [key: string]: string };
 
         include_images?: boolean;
 
+        /**
+         * List of specific sitemap URLs to use for crawling. Only valid when parse_type is
+         * 'sitemap'.
+         */
+        specific_sitemaps?: Array<string>;
+
         use_browser_rendering?: boolean;
       }
 
+      export namespace ParseOptions {
+        export interface ContentSelector {
+          /**
+           * Glob pattern to match against the page URL path. Uses standard glob syntax: \*
+           * matches within a segment, \*\* crosses directories.
+           */
+          path: string;
+
+          /**
+           * CSS selector to extract content from pages matching the path pattern. Supports
+           * standard CSS selectors including class, ID, element, and attribute selectors.
+           */
+          selector: string;
+        }
+      }
+
       export interface StoreOptions {
         storage_id: string;
 
@@ -1014,38 +1367,28 @@ export namespace InstanceListResponse {
 
 export interface InstanceDeleteResponse {
   /**
-   * Use your AI Search ID.
+   * AI Search instance ID. Lowercase alphanumeric, hyphens, and underscores.
    */
   id: string;
 
-  account_id: string;
-
-  account_tag: string;
-
   created_at: string;
 
-  internal_id: string;
-
   modified_at: string;
 
-  source: string;
-
-  token_id: string;
-
-  type: 'r2' | 'web-crawler';
-
-  vectorize_name: string;
-
-  ai_gateway_id?: string;
+  ai_gateway_id?: string | null;
 
   ai_search_model?:
     | '@cf/meta/llama-3.3-70b-instruct-fp8-fast'
+    | '@cf/zai-org/glm-4.7-flash'
     | '@cf/meta/llama-3.1-8b-instruct-fast'
     | '@cf/meta/llama-3.1-8b-instruct-fp8'
     | '@cf/meta/llama-4-scout-17b-16e-instruct'
     | '@cf/qwen/qwen3-30b-a3b-fp8'
     | '@cf/deepseek-ai/deepseek-r1-distill-qwen-32b'
     | '@cf/moonshotai/kimi-k2-instruct'
+    | '@cf/google/gemma-3-12b-it'
+    | '@cf/google/gemma-4-26b-a4b-it'
+    | '@cf/moonshotai/kimi-k2.5'
     | 'anthropic/claude-3-7-sonnet'
     | 'anthropic/claude-sonnet-4'
     | 'anthropic/claude-opus-4'
@@ -1064,62 +1407,86 @@ export interface InstanceDeleteResponse {
     | 'openai/gpt-5'
     | 'openai/gpt-5-mini'
     | 'openai/gpt-5-nano'
-    | '';
+    | ''
+    | null;
 
   cache?: boolean;
 
   cache_threshold?: 'super_strict_match' | 'close_enough' | 'flexible_friend' | 'anything_goes';
 
-  chunk?: boolean;
-
   chunk_overlap?: number;
 
   chunk_size?: number;
 
-  created_by?: string;
+  created_by?: string | null;
+
+  custom_metadata?: Array<InstanceDeleteResponse.CustomMetadata>;
 
   embedding_model?:
+    | '@cf/qwen/qwen3-embedding-0.6b'
     | '@cf/baai/bge-m3'
     | '@cf/baai/bge-large-en-v1.5'
     | '@cf/google/embeddinggemma-300m'
-    | '@cf/qwen/qwen3-embedding-0.6b'
     | 'google-ai-studio/gemini-embedding-001'
+    | 'google-ai-studio/gemini-embedding-2-preview'
     | 'openai/text-embedding-3-small'
     | 'openai/text-embedding-3-large'
-    | '';
+    | ''
+    | null;
 
   enable?: boolean;
 
   engine_version?: number;
 
+  fusion_method?: 'max' | 'rrf';
+
+  /**
+   * @deprecated Deprecated — use index_method instead.
+   */
   hybrid_search_enabled?: boolean;
 
-  last_activity?: string;
+  /**
+   * Controls which storage backends are used during indexing. Defaults to
+   * vector-only.
+   */
+  index_method?: InstanceDeleteResponse.IndexMethod;
+
+  indexing_options?: InstanceDeleteResponse.IndexingOptions | null;
+
+  last_activity?: string | null;
 
   max_num_results?: number;
 
   metadata?: InstanceDeleteResponse.Metadata;
 
-  modified_by?: string;
+  modified_by?: string | null;
+
+  namespace?: string | null;
 
   paused?: boolean;
 
-  public_endpoint_id?: string;
+  public_endpoint_id?: string | null;
 
   public_endpoint_params?: InstanceDeleteResponse.PublicEndpointParams;
 
   reranking?: boolean;
 
-  reranking_model?: '@cf/baai/bge-reranker-base' | '';
+  reranking_model?: '@cf/baai/bge-reranker-base' | '' | null;
+
+  retrieval_options?: InstanceDeleteResponse.RetrievalOptions | null;
 
   rewrite_model?:
     | '@cf/meta/llama-3.3-70b-instruct-fp8-fast'
+    | '@cf/zai-org/glm-4.7-flash'
     | '@cf/meta/llama-3.1-8b-instruct-fast'
     | '@cf/meta/llama-3.1-8b-instruct-fp8'
     | '@cf/meta/llama-4-scout-17b-16e-instruct'
     | '@cf/qwen/qwen3-30b-a3b-fp8'
     | '@cf/deepseek-ai/deepseek-r1-distill-qwen-32b'
     | '@cf/moonshotai/kimi-k2-instruct'
+    | '@cf/google/gemma-3-12b-it'
+    | '@cf/google/gemma-4-26b-a4b-it'
+    | '@cf/moonshotai/kimi-k2.5'
     | 'anthropic/claude-3-7-sonnet'
     | 'anthropic/claude-sonnet-4'
     | 'anthropic/claude-opus-4'
@@ -1138,56 +1505,63 @@ export interface InstanceDeleteResponse {
     | 'openai/gpt-5'
     | 'openai/gpt-5-mini'
     | 'openai/gpt-5-nano'
-    | '';
+    | ''
+    | null;
 
   rewrite_query?: boolean;
 
   score_threshold?: number;
 
-  source_params?: InstanceDeleteResponse.SourceParams;
+  source?: string | null;
+
+  source_params?: InstanceDeleteResponse.SourceParams | null;
 
   status?: string;
 
-  summarization?: boolean;
+  /**
+   * Interval between automatic syncs, in seconds. Allowed values: 900 (15min), 1800
+   * (30min), 3600 (1h), 7200 (2h), 14400 (4h), 21600 (6h), 43200 (12h), 86400 (24h).
+   */
+  sync_interval?: 900 | 1800 | 3600 | 7200 | 14400 | 21600 | 43200 | 86400;
 
-  summarization_model?:
-    | '@cf/meta/llama-3.3-70b-instruct-fp8-fast'
-    | '@cf/meta/llama-3.1-8b-instruct-fast'
-    | '@cf/meta/llama-3.1-8b-instruct-fp8'
-    | '@cf/meta/llama-4-scout-17b-16e-instruct'
-    | '@cf/qwen/qwen3-30b-a3b-fp8'
-    | '@cf/deepseek-ai/deepseek-r1-distill-qwen-32b'
-    | '@cf/moonshotai/kimi-k2-instruct'
-    | 'anthropic/claude-3-7-sonnet'
-    | 'anthropic/claude-sonnet-4'
-    | 'anthropic/claude-opus-4'
-    | 'anthropic/claude-3-5-haiku'
-    | 'cerebras/qwen-3-235b-a22b-instruct'
-    | 'cerebras/qwen-3-235b-a22b-thinking'
-    | 'cerebras/llama-3.3-70b'
-    | 'cerebras/llama-4-maverick-17b-128e-instruct'
-    | 'cerebras/llama-4-scout-17b-16e-instruct'
-    | 'cerebras/gpt-oss-120b'
-    | 'google-ai-studio/gemini-2.5-flash'
-    | 'google-ai-studio/gemini-2.5-pro'
-    | 'grok/grok-4'
-    | 'groq/llama-3.3-70b-versatile'
-    | 'groq/llama-3.1-8b-instant'
-    | 'openai/gpt-5'
-    | 'openai/gpt-5-mini'
-    | 'openai/gpt-5-nano'
-    | '';
+  token_id?: string;
 
-  system_prompt_ai_search?: string;
+  type?: 'r2' | 'web-crawler' | null;
+}
 
-  system_prompt_index_summarization?: string;
+export namespace InstanceDeleteResponse {
+  export interface CustomMetadata {
+    data_type: 'text' | 'number' | 'boolean' | 'datetime';
 
-  system_prompt_rewrite_query?: string;
+    field_name: string;
+  }
 
-  vectorize_active_namespace?: string;
-}
+  /**
+   * Controls which storage backends are used during indexing. Defaults to
+   * vector-only.
+   */
+  export interface IndexMethod {
+    /**
+     * Enable keyword (BM25) storage backend.
+     */
+    keyword: boolean;
+
+    /**
+     * Enable vector (embedding) storage backend.
+     */
+    vector: boolean;
+  }
+
+  export interface IndexingOptions {
+    /**
+     * Tokenizer used for keyword search indexing. porter provides word-level
+     * tokenization with Porter stemming (good for natural language queries). trigram
+     * enables character-level substring matching (good for partial matches, code,
+     * identifiers). Changing this triggers a full re-index. Defaults to porter.
+     */
+    keyword_tokenizer?: 'porter' | 'trigram';
+  }
 
-export namespace InstanceDeleteResponse {
   export interface Metadata {
     created_from_aisearch_wizard?: boolean;
 
@@ -1217,6 +1591,8 @@ export namespace InstanceDeleteResponse {
     }
 
     export interface Mcp {
+      description?: string;
+
       /**
        * Disable MCP endpoint for this public endpoint
        */
@@ -1239,16 +1615,54 @@ export namespace InstanceDeleteResponse {
     }
   }
 
+  export interface RetrievalOptions {
+    /**
+     * Metadata fields to boost search results by. Each entry specifies a metadata
+     * field and an optional direction. Direction defaults to 'asc' for numeric fields
+     * and 'exists' for text/boolean fields. Fields must match 'timestamp' or a defined
+     * custom_metadata field.
+     */
+    boost_by?: Array<RetrievalOptions.BoostBy>;
+
+    /**
+     * Controls which documents are candidates for BM25 scoring. 'and' restricts
+     * candidates to documents containing all query terms; 'or' includes any document
+     * containing at least one term, ranked by BM25 relevance. Defaults to 'and'.
+     */
+    keyword_match_mode?: 'and' | 'or';
+  }
+
+  export namespace RetrievalOptions {
+    export interface BoostBy {
+      /**
+       * Metadata field name to boost by. Use 'timestamp' for document freshness, or any
+       * custom_metadata field. Numeric and datetime fields support asc/desc directions;
+       * text/boolean fields support exists/not_exists.
+       */
+      field: string;
+
+      /**
+       * Boost direction. 'desc' = higher values rank higher (e.g. newer timestamps).
+       * 'asc' = lower values rank higher. 'exists' = boost chunks that have the field.
+       * 'not_exists' = boost chunks that lack the field. Optional ��� defaults to 'asc'
+       * for numeric/datetime fields, 'exists' for text/boolean fields.
+       */
+      direction?: 'asc' | 'desc' | 'exists' | 'not_exists';
+    }
+  }
+
   export interface SourceParams {
     /**
-     * List of path patterns to exclude. Supports wildcards (e.g., _/admin/_,
-     * /private/\*_, _\private\*)
+     * List of path patterns to exclude. Uses micromatch glob syntax: \* matches within
+     * a path segment, ** matches across path segments (e.g., /admin/** matches
+     * /admin/users and /admin/settings/advanced)
      */
     exclude_items?: Array<string>;
 
     /**
-     * List of path patterns to include. Supports wildcards (e.g., _/blog/_.html,
-     * /docs/\*_, _\blog\*.html)
+     * List of path patterns to include. Uses micromatch glob syntax: \* matches within
+     * a path segment, ** matches across path segments (e.g., /blog/** matches
+     * /blog/post and /blog/2024/post)
      */
     include_items?: Array<string>;
 
@@ -1261,22 +1675,65 @@ export namespace InstanceDeleteResponse {
 
   export namespace SourceParams {
     export interface WebCrawler {
+      crawl_options?: WebCrawler.CrawlOptions;
+
       parse_options?: WebCrawler.ParseOptions;
 
-      parse_type?: 'sitemap' | 'feed-rss';
+      parse_type?: 'sitemap' | 'feed-rss' | 'crawl';
 
       store_options?: WebCrawler.StoreOptions;
     }
 
     export namespace WebCrawler {
+      export interface CrawlOptions {
+        depth?: number;
+
+        include_external_links?: boolean;
+
+        include_subdomains?: boolean;
+
+        max_age?: number;
+
+        source?: 'all' | 'sitemaps' | 'links';
+      }
+
       export interface ParseOptions {
+        /**
+         * List of path-to-selector mappings for extracting specific content from crawled
+         * pages. Each entry pairs a URL glob pattern with a CSS selector. The first
+         * matching path wins. Only the matched HTML fragment is stored and indexed.
+         */
+        content_selector?: Array<ParseOptions.ContentSelector>;
+
         include_headers?: { [key: string]: string };
 
         include_images?: boolean;
 
+        /**
+         * List of specific sitemap URLs to use for crawling. Only valid when parse_type is
+         * 'sitemap'.
+         */
+        specific_sitemaps?: Array<string>;
+
         use_browser_rendering?: boolean;
       }
 
+      export namespace ParseOptions {
+        export interface ContentSelector {
+          /**
+           * Glob pattern to match against the page URL path. Uses standard glob syntax: \*
+           * matches within a segment, \*\* crosses directories.
+           */
+          path: string;
+
+          /**
+           * CSS selector to extract content from pages matching the path pattern. Supports
+           * standard CSS selectors including class, ID, element, and attribute selectors.
+           */
+          selector: string;
+        }
+      }
+
       export interface StoreOptions {
         storage_id: string;
 
@@ -1288,40 +1745,100 @@ export namespace InstanceDeleteResponse {
   }
 }
 
-export interface InstanceReadResponse {
-  /**
-   * Use your AI Search ID.
-   */
-  id: string;
+export interface InstanceChatCompletionsResponse {
+  choices: Array<InstanceChatCompletionsResponse.Choice>;
 
-  account_id: string;
+  chunks: Array<InstanceChatCompletionsResponse.Chunk>;
 
-  account_tag: string;
+  id?: string;
 
-  created_at: string;
+  model?: string;
 
-  internal_id: string;
+  object?: string;
 
-  modified_at: string;
+  [k: string]: unknown;
+}
 
-  source: string;
+export namespace InstanceChatCompletionsResponse {
+  export interface Choice {
+    message: Choice.Message;
 
-  token_id: string;
+    index?: number;
+  }
+
+  export namespace Choice {
+    export interface Message {
+      content: string | null;
+
+      role: 'system' | 'developer' | 'user' | 'assistant' | 'tool';
+
+      [k: string]: unknown;
+    }
+  }
+
+  export interface Chunk {
+    id: string;
+
+    score: number;
+
+    text: string;
+
+    type: string;
+
+    item?: Chunk.Item;
+
+    scoring_details?: Chunk.ScoringDetails;
+  }
+
+  export namespace Chunk {
+    export interface Item {
+      key: string;
+
+      metadata?: { [key: string]: unknown };
+
+      timestamp?: number;
+    }
+
+    export interface ScoringDetails {
+      fusion_method?: 'rrf' | 'max';
+
+      keyword_rank?: number;
+
+      keyword_score?: number;
+
+      reranking_score?: number;
+
+      vector_rank?: number;
+
+      vector_score?: number;
+    }
+  }
+}
+
+export interface InstanceReadResponse {
+  /**
+   * AI Search instance ID. Lowercase alphanumeric, hyphens, and underscores.
+   */
+  id: string;
 
-  type: 'r2' | 'web-crawler';
+  created_at: string;
 
-  vectorize_name: string;
+  modified_at: string;
 
-  ai_gateway_id?: string;
+  ai_gateway_id?: string | null;
 
   ai_search_model?:
     | '@cf/meta/llama-3.3-70b-instruct-fp8-fast'
+    | '@cf/zai-org/glm-4.7-flash'
     | '@cf/meta/llama-3.1-8b-instruct-fast'
     | '@cf/meta/llama-3.1-8b-instruct-fp8'
     | '@cf/meta/llama-4-scout-17b-16e-instruct'
     | '@cf/qwen/qwen3-30b-a3b-fp8'
     | '@cf/deepseek-ai/deepseek-r1-distill-qwen-32b'
     | '@cf/moonshotai/kimi-k2-instruct'
+    | '@cf/google/gemma-3-12b-it'
+    | '@cf/google/gemma-4-26b-a4b-it'
+    | '@cf/moonshotai/kimi-k2.5'
     | 'anthropic/claude-3-7-sonnet'
     | 'anthropic/claude-sonnet-4'
     | 'anthropic/claude-opus-4'
@@ -1340,62 +1857,86 @@ export interface InstanceReadResponse {
     | 'openai/gpt-5'
     | 'openai/gpt-5-mini'
     | 'openai/gpt-5-nano'
-    | '';
+    | ''
+    | null;
 
   cache?: boolean;
 
   cache_threshold?: 'super_strict_match' | 'close_enough' | 'flexible_friend' | 'anything_goes';
 
-  chunk?: boolean;
-
   chunk_overlap?: number;
 
   chunk_size?: number;
 
-  created_by?: string;
+  created_by?: string | null;
+
+  custom_metadata?: Array<InstanceReadResponse.CustomMetadata>;
 
   embedding_model?:
+    | '@cf/qwen/qwen3-embedding-0.6b'
     | '@cf/baai/bge-m3'
     | '@cf/baai/bge-large-en-v1.5'
     | '@cf/google/embeddinggemma-300m'
-    | '@cf/qwen/qwen3-embedding-0.6b'
     | 'google-ai-studio/gemini-embedding-001'
+    | 'google-ai-studio/gemini-embedding-2-preview'
     | 'openai/text-embedding-3-small'
     | 'openai/text-embedding-3-large'
-    | '';
+    | ''
+    | null;
 
   enable?: boolean;
 
   engine_version?: number;
 
+  fusion_method?: 'max' | 'rrf';
+
+  /**
+   * @deprecated Deprecated — use index_method instead.
+   */
   hybrid_search_enabled?: boolean;
 
-  last_activity?: string;
+  /**
+   * Controls which storage backends are used during indexing. Defaults to
+   * vector-only.
+   */
+  index_method?: InstanceReadResponse.IndexMethod;
+
+  indexing_options?: InstanceReadResponse.IndexingOptions | null;
+
+  last_activity?: string | null;
 
   max_num_results?: number;
 
   metadata?: InstanceReadResponse.Metadata;
 
-  modified_by?: string;
+  modified_by?: string | null;
+
+  namespace?: string | null;
 
   paused?: boolean;
 
-  public_endpoint_id?: string;
+  public_endpoint_id?: string | null;
 
   public_endpoint_params?: InstanceReadResponse.PublicEndpointParams;
 
   reranking?: boolean;
 
-  reranking_model?: '@cf/baai/bge-reranker-base' | '';
+  reranking_model?: '@cf/baai/bge-reranker-base' | '' | null;
+
+  retrieval_options?: InstanceReadResponse.RetrievalOptions | null;
 
   rewrite_model?:
     | '@cf/meta/llama-3.3-70b-instruct-fp8-fast'
+    | '@cf/zai-org/glm-4.7-flash'
     | '@cf/meta/llama-3.1-8b-instruct-fast'
     | '@cf/meta/llama-3.1-8b-instruct-fp8'
     | '@cf/meta/llama-4-scout-17b-16e-instruct'
     | '@cf/qwen/qwen3-30b-a3b-fp8'
     | '@cf/deepseek-ai/deepseek-r1-distill-qwen-32b'
     | '@cf/moonshotai/kimi-k2-instruct'
+    | '@cf/google/gemma-3-12b-it'
+    | '@cf/google/gemma-4-26b-a4b-it'
+    | '@cf/moonshotai/kimi-k2.5'
     | 'anthropic/claude-3-7-sonnet'
     | 'anthropic/claude-sonnet-4'
     | 'anthropic/claude-opus-4'
@@ -1414,56 +1955,63 @@ export interface InstanceReadResponse {
     | 'openai/gpt-5'
     | 'openai/gpt-5-mini'
     | 'openai/gpt-5-nano'
-    | '';
+    | ''
+    | null;
 
   rewrite_query?: boolean;
 
   score_threshold?: number;
 
-  source_params?: InstanceReadResponse.SourceParams;
+  source?: string | null;
+
+  source_params?: InstanceReadResponse.SourceParams | null;
 
   status?: string;
 
-  summarization?: boolean;
+  /**
+   * Interval between automatic syncs, in seconds. Allowed values: 900 (15min), 1800
+   * (30min), 3600 (1h), 7200 (2h), 14400 (4h), 21600 (6h), 43200 (12h), 86400 (24h).
+   */
+  sync_interval?: 900 | 1800 | 3600 | 7200 | 14400 | 21600 | 43200 | 86400;
 
-  summarization_model?:
-    | '@cf/meta/llama-3.3-70b-instruct-fp8-fast'
-    | '@cf/meta/llama-3.1-8b-instruct-fast'
-    | '@cf/meta/llama-3.1-8b-instruct-fp8'
-    | '@cf/meta/llama-4-scout-17b-16e-instruct'
-    | '@cf/qwen/qwen3-30b-a3b-fp8'
-    | '@cf/deepseek-ai/deepseek-r1-distill-qwen-32b'
-    | '@cf/moonshotai/kimi-k2-instruct'
-    | 'anthropic/claude-3-7-sonnet'
-    | 'anthropic/claude-sonnet-4'
-    | 'anthropic/claude-opus-4'
-    | 'anthropic/claude-3-5-haiku'
-    | 'cerebras/qwen-3-235b-a22b-instruct'
-    | 'cerebras/qwen-3-235b-a22b-thinking'
-    | 'cerebras/llama-3.3-70b'
-    | 'cerebras/llama-4-maverick-17b-128e-instruct'
-    | 'cerebras/llama-4-scout-17b-16e-instruct'
-    | 'cerebras/gpt-oss-120b'
-    | 'google-ai-studio/gemini-2.5-flash'
-    | 'google-ai-studio/gemini-2.5-pro'
-    | 'grok/grok-4'
-    | 'groq/llama-3.3-70b-versatile'
-    | 'groq/llama-3.1-8b-instant'
-    | 'openai/gpt-5'
-    | 'openai/gpt-5-mini'
-    | 'openai/gpt-5-nano'
-    | '';
+  token_id?: string;
 
-  system_prompt_ai_search?: string;
+  type?: 'r2' | 'web-crawler' | null;
+}
 
-  system_prompt_index_summarization?: string;
+export namespace InstanceReadResponse {
+  export interface CustomMetadata {
+    data_type: 'text' | 'number' | 'boolean' | 'datetime';
 
-  system_prompt_rewrite_query?: string;
+    field_name: string;
+  }
 
-  vectorize_active_namespace?: string;
-}
+  /**
+   * Controls which storage backends are used during indexing. Defaults to
+   * vector-only.
+   */
+  export interface IndexMethod {
+    /**
+     * Enable keyword (BM25) storage backend.
+     */
+    keyword: boolean;
+
+    /**
+     * Enable vector (embedding) storage backend.
+     */
+    vector: boolean;
+  }
+
+  export interface IndexingOptions {
+    /**
+     * Tokenizer used for keyword search indexing. porter provides word-level
+     * tokenization with Porter stemming (good for natural language queries). trigram
+     * enables character-level substring matching (good for partial matches, code,
+     * identifiers). Changing this triggers a full re-index. Defaults to porter.
+     */
+    keyword_tokenizer?: 'porter' | 'trigram';
+  }
 
-export namespace InstanceReadResponse {
   export interface Metadata {
     created_from_aisearch_wizard?: boolean;
 
@@ -1493,6 +2041,8 @@ export namespace InstanceReadResponse {
     }
 
     export interface Mcp {
+      description?: string;
+
       /**
        * Disable MCP endpoint for this public endpoint
        */
@@ -1515,16 +2065,54 @@ export namespace InstanceReadResponse {
     }
   }
 
+  export interface RetrievalOptions {
+    /**
+     * Metadata fields to boost search results by. Each entry specifies a metadata
+     * field and an optional direction. Direction defaults to 'asc' for numeric fields
+     * and 'exists' for text/boolean fields. Fields must match 'timestamp' or a defined
+     * custom_metadata field.
+     */
+    boost_by?: Array<RetrievalOptions.BoostBy>;
+
+    /**
+     * Controls which documents are candidates for BM25 scoring. 'and' restricts
+     * candidates to documents containing all query terms; 'or' includes any document
+     * containing at least one term, ranked by BM25 relevance. Defaults to 'and'.
+     */
+    keyword_match_mode?: 'and' | 'or';
+  }
+
+  export namespace RetrievalOptions {
+    export interface BoostBy {
+      /**
+       * Metadata field name to boost by. Use 'timestamp' for document freshness, or any
+       * custom_metadata field. Numeric and datetime fields support asc/desc directions;
+       * text/boolean fields support exists/not_exists.
+       */
+      field: string;
+
+      /**
+       * Boost direction. 'desc' = higher values rank higher (e.g. newer timestamps).
+       * 'asc' = lower values rank higher. 'exists' = boost chunks that have the field.
+       * 'not_exists' = boost chunks that lack the field. Optional ��� defaults to 'asc'
+       * for numeric/datetime fields, 'exists' for text/boolean fields.
+       */
+      direction?: 'asc' | 'desc' | 'exists' | 'not_exists';
+    }
+  }
+
   export interface SourceParams {
     /**
-     * List of path patterns to exclude. Supports wildcards (e.g., _/admin/_,
-     * /private/\*_, _\private\*)
+     * List of path patterns to exclude. Uses micromatch glob syntax: \* matches within
+     * a path segment, ** matches across path segments (e.g., /admin/** matches
+     * /admin/users and /admin/settings/advanced)
      */
     exclude_items?: Array<string>;
 
     /**
-     * List of path patterns to include. Supports wildcards (e.g., _/blog/_.html,
-     * /docs/\*_, _\blog\*.html)
+     * List of path patterns to include. Uses micromatch glob syntax: \* matches within
+     * a path segment, ** matches across path segments (e.g., /blog/** matches
+     * /blog/post and /blog/2024/post)
      */
     include_items?: Array<string>;
 
@@ -1537,22 +2125,65 @@ export namespace InstanceReadResponse {
 
   export namespace SourceParams {
     export interface WebCrawler {
+      crawl_options?: WebCrawler.CrawlOptions;
+
       parse_options?: WebCrawler.ParseOptions;
 
-      parse_type?: 'sitemap' | 'feed-rss';
+      parse_type?: 'sitemap' | 'feed-rss' | 'crawl';
 
       store_options?: WebCrawler.StoreOptions;
     }
 
     export namespace WebCrawler {
+      export interface CrawlOptions {
+        depth?: number;
+
+        include_external_links?: boolean;
+
+        include_subdomains?: boolean;
+
+        max_age?: number;
+
+        source?: 'all' | 'sitemaps' | 'links';
+      }
+
       export interface ParseOptions {
+        /**
+         * List of path-to-selector mappings for extracting specific content from crawled
+         * pages. Each entry pairs a URL glob pattern with a CSS selector. The first
+         * matching path wins. Only the matched HTML fragment is stored and indexed.
+         */
+        content_selector?: Array<ParseOptions.ContentSelector>;
+
         include_headers?: { [key: string]: string };
 
         include_images?: boolean;
 
+        /**
+         * List of specific sitemap URLs to use for crawling. Only valid when parse_type is
+         * 'sitemap'.
+         */
+        specific_sitemaps?: Array<string>;
+
         use_browser_rendering?: boolean;
       }
 
+      export namespace ParseOptions {
+        export interface ContentSelector {
+          /**
+           * Glob pattern to match against the page URL path. Uses standard glob syntax: \*
+           * matches within a segment, \*\* crosses directories.
+           */
+          path: string;
+
+          /**
+           * CSS selector to extract content from pages matching the path pattern. Supports
+           * standard CSS selectors including class, ID, element, and attribute selectors.
+           */
+          selector: string;
+        }
+      }
+
       export interface StoreOptions {
         storage_id: string;
 
@@ -1564,9 +2195,60 @@ export namespace InstanceReadResponse {
   }
 }
 
+export interface InstanceSearchResponse {
+  chunks: Array<InstanceSearchResponse.Chunk>;
+
+  search_query: string;
+}
+
+export namespace InstanceSearchResponse {
+  export interface Chunk {
+    id: string;
+
+    score: number;
+
+    text: string;
+
+    type: string;
+
+    item?: Chunk.Item;
+
+    scoring_details?: Chunk.ScoringDetails;
+  }
+
+  export namespace Chunk {
+    export interface Item {
+      key: string;
+
+      metadata?: { [key: string]: unknown };
+
+      timestamp?: number;
+    }
+
+    export interface ScoringDetails {
+      fusion_method?: 'rrf' | 'max';
+
+      keyword_rank?: number;
+
+      keyword_score?: number;
+
+      reranking_score?: number;
+
+      vector_rank?: number;
+
+      vector_score?: number;
+    }
+  }
+}
+
 export interface InstanceStatsResponse {
   completed?: number;
 
+  /**
+   * Engine-specific metadata. Present only for managed (v3) instances.
+   */
+  engine?: InstanceStatsResponse.Engine;
+
   error?: number;
 
   file_embed_errors?: { [key: string]: unknown };
@@ -1575,6 +2257,8 @@ export interface InstanceStatsResponse {
 
   last_activity?: string;
 
+  outdated?: number;
+
   queued?: number;
 
   running?: number;
@@ -1582,48 +2266,77 @@ export interface InstanceStatsResponse {
   skipped?: number;
 }
 
-export interface InstanceCreateParams {
+export namespace InstanceStatsResponse {
   /**
-   * Path param:
+   * Engine-specific metadata. Present only for managed (v3) instances.
    */
-  account_id: string;
+  export interface Engine {
+    /**
+     * R2 bucket storage usage in bytes.
+     */
+    r2?: Engine.R2;
 
-  /**
-   * Body param: Use your AI Search ID.
-   */
-  id: string;
+    /**
+     * Vectorize index metadata (dimensions, vector count).
+     */
+    vectorize?: Engine.Vectorize;
+  }
 
-  /**
-   * Body param:
-   */
-  source: string;
+  export namespace Engine {
+    /**
+     * R2 bucket storage usage in bytes.
+     */
+    export interface R2 {
+      metadataSizeBytes: number;
+
+      objectCount: number;
+
+      payloadSizeBytes: number;
+    }
+
+    /**
+     * Vectorize index metadata (dimensions, vector count).
+     */
+    export interface Vectorize {
+      dimensions: number;
 
+      vectorsCount: number;
+    }
+  }
+}
+
+export interface InstanceCreateParams {
   /**
-   * Body param:
+   * Path param
    */
-  token_id: string;
+  account_id: string;
 
   /**
-   * Body param:
+   * Body param: AI Search instance ID. Lowercase alphanumeric, hyphens, and
+   * underscores.
    */
-  type: 'r2' | 'web-crawler';
+  id: string;
 
   /**
-   * Body param:
+   * Body param
    */
-  ai_gateway_id?: string;
+  ai_gateway_id?: string | null;
 
   /**
-   * Body param:
+   * Body param
    */
   ai_search_model?:
     | '@cf/meta/llama-3.3-70b-instruct-fp8-fast'
+    | '@cf/zai-org/glm-4.7-flash'
     | '@cf/meta/llama-3.1-8b-instruct-fast'
     | '@cf/meta/llama-3.1-8b-instruct-fp8'
     | '@cf/meta/llama-4-scout-17b-16e-instruct'
     | '@cf/qwen/qwen3-30b-a3b-fp8'
     | '@cf/deepseek-ai/deepseek-r1-distill-qwen-32b'
     | '@cf/moonshotai/kimi-k2-instruct'
+    | '@cf/google/gemma-3-12b-it'
+    | '@cf/google/gemma-4-26b-a4b-it'
+    | '@cf/moonshotai/kimi-k2.5'
     | 'anthropic/claude-3-7-sonnet'
     | 'anthropic/claude-sonnet-4'
     | 'anthropic/claude-opus-4'
@@ -1642,77 +2355,120 @@ export interface InstanceCreateParams {
     | 'openai/gpt-5'
     | 'openai/gpt-5-mini'
     | 'openai/gpt-5-nano'
-    | '';
+    | ''
+    | null;
 
   /**
-   * Body param:
+   * Body param
+   */
+  cache?: boolean;
+
+  /**
+   * Body param
+   */
+  cache_threshold?: 'super_strict_match' | 'close_enough' | 'flexible_friend' | 'anything_goes';
+
+  /**
+   * Body param
    */
   chunk?: boolean;
 
   /**
-   * Body param:
+   * Body param
    */
   chunk_overlap?: number;
 
   /**
-   * Body param:
+   * Body param
    */
   chunk_size?: number;
 
   /**
-   * Body param:
+   * Body param
+   */
+  custom_metadata?: Array<InstanceCreateParams.CustomMetadata>;
+
+  /**
+   * Body param
    */
   embedding_model?:
+    | '@cf/qwen/qwen3-embedding-0.6b'
     | '@cf/baai/bge-m3'
     | '@cf/baai/bge-large-en-v1.5'
     | '@cf/google/embeddinggemma-300m'
-    | '@cf/qwen/qwen3-embedding-0.6b'
     | 'google-ai-studio/gemini-embedding-001'
+    | 'google-ai-studio/gemini-embedding-2-preview'
     | 'openai/text-embedding-3-small'
     | 'openai/text-embedding-3-large'
-    | '';
+    | ''
+    | null;
 
   /**
-   * Body param:
+   * Body param
+   */
+  fusion_method?: 'max' | 'rrf';
+
+  /**
+   * @deprecated Body param: Deprecated — use index_method instead.
    */
   hybrid_search_enabled?: boolean;
 
   /**
-   * Body param:
+   * Body param: Controls which storage backends are used during indexing. Defaults
+   * to vector-only.
+   */
+  index_method?: InstanceCreateParams.IndexMethod;
+
+  /**
+   * Body param
+   */
+  indexing_options?: InstanceCreateParams.IndexingOptions | null;
+
+  /**
+   * Body param
    */
   max_num_results?: number;
 
   /**
-   * Body param:
+   * Body param
    */
   metadata?: InstanceCreateParams.Metadata;
 
   /**
-   * Body param:
+   * Body param
    */
   public_endpoint_params?: InstanceCreateParams.PublicEndpointParams;
 
   /**
-   * Body param:
+   * Body param
    */
   reranking?: boolean;
 
   /**
-   * Body param:
+   * Body param
    */
-  reranking_model?: '@cf/baai/bge-reranker-base' | '';
+  reranking_model?: '@cf/baai/bge-reranker-base' | '' | null;
 
   /**
-   * Body param:
+   * Body param
+   */
+  retrieval_options?: InstanceCreateParams.RetrievalOptions | null;
+
+  /**
+   * Body param
    */
   rewrite_model?:
     | '@cf/meta/llama-3.3-70b-instruct-fp8-fast'
+    | '@cf/zai-org/glm-4.7-flash'
     | '@cf/meta/llama-3.1-8b-instruct-fast'
     | '@cf/meta/llama-3.1-8b-instruct-fp8'
     | '@cf/meta/llama-4-scout-17b-16e-instruct'
     | '@cf/qwen/qwen3-30b-a3b-fp8'
     | '@cf/deepseek-ai/deepseek-r1-distill-qwen-32b'
     | '@cf/moonshotai/kimi-k2-instruct'
+    | '@cf/google/gemma-3-12b-it'
+    | '@cf/google/gemma-4-26b-a4b-it'
+    | '@cf/moonshotai/kimi-k2.5'
     | 'anthropic/claude-3-7-sonnet'
     | 'anthropic/claude-sonnet-4'
     | 'anthropic/claude-opus-4'
@@ -1731,25 +2487,80 @@ export interface InstanceCreateParams {
     | 'openai/gpt-5'
     | 'openai/gpt-5-mini'
     | 'openai/gpt-5-nano'
-    | '';
+    | ''
+    | null;
 
   /**
-   * Body param:
+   * Body param
    */
   rewrite_query?: boolean;
 
   /**
-   * Body param:
+   * Body param
    */
   score_threshold?: number;
 
   /**
-   * Body param:
+   * Body param
+   */
+  source?: string | null;
+
+  /**
+   * Body param
+   */
+  source_params?: InstanceCreateParams.SourceParams | null;
+
+  /**
+   * Body param: Interval between automatic syncs, in seconds. Allowed values: 900
+   * (15min), 1800 (30min), 3600 (1h), 7200 (2h), 14400 (4h), 21600 (6h), 43200
+   * (12h), 86400 (24h).
+   */
+  sync_interval?: 900 | 1800 | 3600 | 7200 | 14400 | 21600 | 43200 | 86400;
+
+  /**
+   * Body param
    */
-  source_params?: InstanceCreateParams.SourceParams;
+  token_id?: string;
+
+  /**
+   * Body param
+   */
+  type?: 'r2' | 'web-crawler' | null;
 }
 
 export namespace InstanceCreateParams {
+  export interface CustomMetadata {
+    data_type: 'text' | 'number' | 'boolean' | 'datetime';
+
+    field_name: string;
+  }
+
+  /**
+   * Controls which storage backends are used during indexing. Defaults to
+   * vector-only.
+   */
+  export interface IndexMethod {
+    /**
+     * Enable keyword (BM25) storage backend.
+     */
+    keyword: boolean;
+
+    /**
+     * Enable vector (embedding) storage backend.
+     */
+    vector: boolean;
+  }
+
+  export interface IndexingOptions {
+    /**
+     * Tokenizer used for keyword search indexing. porter provides word-level
+     * tokenization with Porter stemming (good for natural language queries). trigram
+     * enables character-level substring matching (good for partial matches, code,
+     * identifiers). Changing this triggers a full re-index. Defaults to porter.
+     */
+    keyword_tokenizer?: 'porter' | 'trigram';
+  }
+
   export interface Metadata {
     created_from_aisearch_wizard?: boolean;
 
@@ -1779,6 +2590,8 @@ export namespace InstanceCreateParams {
     }
 
     export interface Mcp {
+      description?: string;
+
       /**
        * Disable MCP endpoint for this public endpoint
        */
@@ -1801,16 +2614,54 @@ export namespace InstanceCreateParams {
     }
   }
 
+  export interface RetrievalOptions {
+    /**
+     * Metadata fields to boost search results by. Each entry specifies a metadata
+     * field and an optional direction. Direction defaults to 'asc' for numeric fields
+     * and 'exists' for text/boolean fields. Fields must match 'timestamp' or a defined
+     * custom_metadata field.
+     */
+    boost_by?: Array<RetrievalOptions.BoostBy>;
+
+    /**
+     * Controls which documents are candidates for BM25 scoring. 'and' restricts
+     * candidates to documents containing all query terms; 'or' includes any document
+     * containing at least one term, ranked by BM25 relevance. Defaults to 'and'.
+     */
+    keyword_match_mode?: 'and' | 'or';
+  }
+
+  export namespace RetrievalOptions {
+    export interface BoostBy {
+      /**
+       * Metadata field name to boost by. Use 'timestamp' for document freshness, or any
+       * custom_metadata field. Numeric and datetime fields support asc/desc directions;
+       * text/boolean fields support exists/not_exists.
+       */
+      field: string;
+
+      /**
+       * Boost direction. 'desc' = higher values rank higher (e.g. newer timestamps).
+       * 'asc' = lower values rank higher. 'exists' = boost chunks that have the field.
+       * 'not_exists' = boost chunks that lack the field. Optional ��� defaults to 'asc'
+       * for numeric/datetime fields, 'exists' for text/boolean fields.
+       */
+      direction?: 'asc' | 'desc' | 'exists' | 'not_exists';
+    }
+  }
+
   export interface SourceParams {
     /**
-     * List of path patterns to exclude. Supports wildcards (e.g., _/admin/_,
-     * /private/\*_, _\private\*)
+     * List of path patterns to exclude. Uses micromatch glob syntax: \* matches within
+     * a path segment, ** matches across path segments (e.g., /admin/** matches
+     * /admin/users and /admin/settings/advanced)
      */
     exclude_items?: Array<string>;
 
     /**
-     * List of path patterns to include. Supports wildcards (e.g., _/blog/_.html,
-     * /docs/\*_, _\blog\*.html)
+     * List of path patterns to include. Uses micromatch glob syntax: \* matches within
+     * a path segment, ** matches across path segments (e.g., /blog/** matches
+     * /blog/post and /blog/2024/post)
      */
     include_items?: Array<string>;
 
@@ -1823,22 +2674,65 @@ export namespace InstanceCreateParams {
 
   export namespace SourceParams {
     export interface WebCrawler {
+      crawl_options?: WebCrawler.CrawlOptions;
+
       parse_options?: WebCrawler.ParseOptions;
 
-      parse_type?: 'sitemap' | 'feed-rss';
+      parse_type?: 'sitemap' | 'feed-rss' | 'crawl';
 
       store_options?: WebCrawler.StoreOptions;
     }
 
     export namespace WebCrawler {
+      export interface CrawlOptions {
+        depth?: number;
+
+        include_external_links?: boolean;
+
+        include_subdomains?: boolean;
+
+        max_age?: number;
+
+        source?: 'all' | 'sitemaps' | 'links';
+      }
+
       export interface ParseOptions {
+        /**
+         * List of path-to-selector mappings for extracting specific content from crawled
+         * pages. Each entry pairs a URL glob pattern with a CSS selector. The first
+         * matching path wins. Only the matched HTML fragment is stored and indexed.
+         */
+        content_selector?: Array<ParseOptions.ContentSelector>;
+
         include_headers?: { [key: string]: string };
 
         include_images?: boolean;
 
+        /**
+         * List of specific sitemap URLs to use for crawling. Only valid when parse_type is
+         * 'sitemap'.
+         */
+        specific_sitemaps?: Array<string>;
+
         use_browser_rendering?: boolean;
       }
 
+      export namespace ParseOptions {
+        export interface ContentSelector {
+          /**
+           * Glob pattern to match against the page URL path. Uses standard glob syntax: \*
+           * matches within a segment, \*\* crosses directories.
+           */
+          path: string;
+
+          /**
+           * CSS selector to extract content from pages matching the path pattern. Supports
+           * standard CSS selectors including class, ID, element, and attribute selectors.
+           */
+          selector: string;
+        }
+      }
+
       export interface StoreOptions {
         storage_id: string;
 
@@ -1852,26 +2746,30 @@ export namespace InstanceCreateParams {
 
 export interface InstanceUpdateParams {
   /**
-   * Path param:
+   * Path param
    */
   account_id: string;
 
   /**
-   * Body param:
+   * Body param
    */
-  ai_gateway_id?: string;
+  ai_gateway_id?: string | null;
 
   /**
-   * Body param:
+   * Body param
    */
   ai_search_model?:
     | '@cf/meta/llama-3.3-70b-instruct-fp8-fast'
+    | '@cf/zai-org/glm-4.7-flash'
     | '@cf/meta/llama-3.1-8b-instruct-fast'
     | '@cf/meta/llama-3.1-8b-instruct-fp8'
     | '@cf/meta/llama-4-scout-17b-16e-instruct'
     | '@cf/qwen/qwen3-30b-a3b-fp8'
     | '@cf/deepseek-ai/deepseek-r1-distill-qwen-32b'
     | '@cf/moonshotai/kimi-k2-instruct'
+    | '@cf/google/gemma-3-12b-it'
+    | '@cf/google/gemma-4-26b-a4b-it'
+    | '@cf/moonshotai/kimi-k2.5'
     | 'anthropic/claude-3-7-sonnet'
     | 'anthropic/claude-sonnet-4'
     | 'anthropic/claude-opus-4'
@@ -1890,92 +2788,120 @@ export interface InstanceUpdateParams {
     | 'openai/gpt-5'
     | 'openai/gpt-5-mini'
     | 'openai/gpt-5-nano'
-    | '';
+    | ''
+    | null;
 
   /**
-   * Body param:
+   * Body param
    */
   cache?: boolean;
 
   /**
-   * Body param:
+   * Body param
    */
   cache_threshold?: 'super_strict_match' | 'close_enough' | 'flexible_friend' | 'anything_goes';
 
   /**
-   * Body param:
+   * Body param
    */
   chunk?: boolean;
 
   /**
-   * Body param:
+   * Body param
    */
   chunk_overlap?: number;
 
   /**
-   * Body param:
+   * Body param
    */
   chunk_size?: number;
 
   /**
-   * Body param:
+   * Body param
+   */
+  custom_metadata?: Array<InstanceUpdateParams.CustomMetadata>;
+
+  /**
+   * Body param
    */
   embedding_model?:
+    | '@cf/qwen/qwen3-embedding-0.6b'
     | '@cf/baai/bge-m3'
     | '@cf/baai/bge-large-en-v1.5'
     | '@cf/google/embeddinggemma-300m'
-    | '@cf/qwen/qwen3-embedding-0.6b'
     | 'google-ai-studio/gemini-embedding-001'
+    | 'google-ai-studio/gemini-embedding-2-preview'
     | 'openai/text-embedding-3-small'
     | 'openai/text-embedding-3-large'
-    | '';
+    | ''
+    | null;
 
   /**
-   * Body param:
+   * Body param
    */
-  hybrid_search_enabled?: boolean;
+  fusion_method?: 'max' | 'rrf';
+
+  /**
+   * Body param: Controls which storage backends are used during indexing. Defaults
+   * to vector-only.
+   */
+  index_method?: InstanceUpdateParams.IndexMethod;
 
   /**
-   * Body param:
+   * Body param
+   */
+  indexing_options?: InstanceUpdateParams.IndexingOptions | null;
+
+  /**
+   * Body param
    */
   max_num_results?: number;
 
   /**
-   * Body param:
+   * Body param
    */
   metadata?: InstanceUpdateParams.Metadata;
 
   /**
-   * Body param:
+   * Body param
    */
   paused?: boolean;
 
   /**
-   * Body param:
+   * Body param
    */
   public_endpoint_params?: InstanceUpdateParams.PublicEndpointParams;
 
   /**
-   * Body param:
+   * Body param
    */
   reranking?: boolean;
 
   /**
-   * Body param:
+   * Body param
+   */
+  reranking_model?: '@cf/baai/bge-reranker-base' | '' | null;
+
+  /**
+   * Body param
    */
-  reranking_model?: '@cf/baai/bge-reranker-base' | '';
+  retrieval_options?: InstanceUpdateParams.RetrievalOptions | null;
 
   /**
-   * Body param:
+   * Body param
    */
   rewrite_model?:
     | '@cf/meta/llama-3.3-70b-instruct-fp8-fast'
+    | '@cf/zai-org/glm-4.7-flash'
     | '@cf/meta/llama-3.1-8b-instruct-fast'
     | '@cf/meta/llama-3.1-8b-instruct-fp8'
     | '@cf/meta/llama-4-scout-17b-16e-instruct'
     | '@cf/qwen/qwen3-30b-a3b-fp8'
     | '@cf/deepseek-ai/deepseek-r1-distill-qwen-32b'
     | '@cf/moonshotai/kimi-k2-instruct'
+    | '@cf/google/gemma-3-12b-it'
+    | '@cf/google/gemma-4-26b-a4b-it'
+    | '@cf/moonshotai/kimi-k2.5'
     | 'anthropic/claude-3-7-sonnet'
     | 'anthropic/claude-sonnet-4'
     | 'anthropic/claude-opus-4'
@@ -1994,39 +2920,44 @@ export interface InstanceUpdateParams {
     | 'openai/gpt-5'
     | 'openai/gpt-5-mini'
     | 'openai/gpt-5-nano'
-    | '';
+    | ''
+    | null;
 
   /**
-   * Body param:
+   * Body param
    */
   rewrite_query?: boolean;
 
   /**
-   * Body param:
+   * Body param
    */
   score_threshold?: number;
 
   /**
-   * Body param:
+   * Body param
    */
-  source_params?: InstanceUpdateParams.SourceParams;
+  source_params?: InstanceUpdateParams.SourceParams | null;
 
   /**
-   * Body param:
+   * Body param
    */
   summarization?: boolean;
 
   /**
-   * Body param:
+   * Body param
    */
   summarization_model?:
     | '@cf/meta/llama-3.3-70b-instruct-fp8-fast'
+    | '@cf/zai-org/glm-4.7-flash'
     | '@cf/meta/llama-3.1-8b-instruct-fast'
     | '@cf/meta/llama-3.1-8b-instruct-fp8'
     | '@cf/meta/llama-4-scout-17b-16e-instruct'
     | '@cf/qwen/qwen3-30b-a3b-fp8'
     | '@cf/deepseek-ai/deepseek-r1-distill-qwen-32b'
     | '@cf/moonshotai/kimi-k2-instruct'
+    | '@cf/google/gemma-3-12b-it'
+    | '@cf/google/gemma-4-26b-a4b-it'
+    | '@cf/moonshotai/kimi-k2.5'
     | 'anthropic/claude-3-7-sonnet'
     | 'anthropic/claude-sonnet-4'
     | 'anthropic/claude-opus-4'
@@ -2045,30 +2976,70 @@ export interface InstanceUpdateParams {
     | 'openai/gpt-5'
     | 'openai/gpt-5-mini'
     | 'openai/gpt-5-nano'
-    | '';
+    | ''
+    | null;
 
   /**
-   * Body param:
+   * Body param: Interval between automatic syncs, in seconds. Allowed values: 900
+   * (15min), 1800 (30min), 3600 (1h), 7200 (2h), 14400 (4h), 21600 (6h), 43200
+   * (12h), 86400 (24h).
    */
-  system_prompt_ai_search?: string;
+  sync_interval?: 900 | 1800 | 3600 | 7200 | 14400 | 21600 | 43200 | 86400;
 
   /**
-   * Body param:
+   * Body param
    */
-  system_prompt_index_summarization?: string;
+  system_prompt_ai_search?: string | null;
 
   /**
-   * Body param:
+   * Body param
    */
-  system_prompt_rewrite_query?: string;
+  system_prompt_index_summarization?: string | null;
 
   /**
-   * Body param:
+   * Body param
+   */
+  system_prompt_rewrite_query?: string | null;
+
+  /**
+   * Body param
    */
   token_id?: string;
 }
 
 export namespace InstanceUpdateParams {
+  export interface CustomMetadata {
+    data_type: 'text' | 'number' | 'boolean' | 'datetime';
+
+    field_name: string;
+  }
+
+  /**
+   * Controls which storage backends are used during indexing. Defaults to
+   * vector-only.
+   */
+  export interface IndexMethod {
+    /**
+     * Enable keyword (BM25) storage backend.
+     */
+    keyword: boolean;
+
+    /**
+     * Enable vector (embedding) storage backend.
+     */
+    vector: boolean;
+  }
+
+  export interface IndexingOptions {
+    /**
+     * Tokenizer used for keyword search indexing. porter provides word-level
+     * tokenization with Porter stemming (good for natural language queries). trigram
+     * enables character-level substring matching (good for partial matches, code,
+     * identifiers). Changing this triggers a full re-index. Defaults to porter.
+     */
+    keyword_tokenizer?: 'porter' | 'trigram';
+  }
+
   export interface Metadata {
     created_from_aisearch_wizard?: boolean;
 
@@ -2098,6 +3069,8 @@ export namespace InstanceUpdateParams {
     }
 
     export interface Mcp {
+      description?: string;
+
       /**
        * Disable MCP endpoint for this public endpoint
        */
@@ -2120,16 +3093,54 @@ export namespace InstanceUpdateParams {
     }
   }
 
+  export interface RetrievalOptions {
+    /**
+     * Metadata fields to boost search results by. Each entry specifies a metadata
+     * field and an optional direction. Direction defaults to 'asc' for numeric fields
+     * and 'exists' for text/boolean fields. Fields must match 'timestamp' or a defined
+     * custom_metadata field.
+     */
+    boost_by?: Array<RetrievalOptions.BoostBy>;
+
+    /**
+     * Controls which documents are candidates for BM25 scoring. 'and' restricts
+     * candidates to documents containing all query terms; 'or' includes any document
+     * containing at least one term, ranked by BM25 relevance. Defaults to 'and'.
+     */
+    keyword_match_mode?: 'and' | 'or';
+  }
+
+  export namespace RetrievalOptions {
+    export interface BoostBy {
+      /**
+       * Metadata field name to boost by. Use 'timestamp' for document freshness, or any
+       * custom_metadata field. Numeric and datetime fields support asc/desc directions;
+       * text/boolean fields support exists/not_exists.
+       */
+      field: string;
+
+      /**
+       * Boost direction. 'desc' = higher values rank higher (e.g. newer timestamps).
+       * 'asc' = lower values rank higher. 'exists' = boost chunks that have the field.
+       * 'not_exists' = boost chunks that lack the field. Optional ��� defaults to 'asc'
+       * for numeric/datetime fields, 'exists' for text/boolean fields.
+       */
+      direction?: 'asc' | 'desc' | 'exists' | 'not_exists';
+    }
+  }
+
   export interface SourceParams {
     /**
-     * List of path patterns to exclude. Supports wildcards (e.g., _/admin/_,
-     * /private/\*_, _\private\*)
+     * List of path patterns to exclude. Uses micromatch glob syntax: \* matches within
+     * a path segment, ** matches across path segments (e.g., /admin/** matches
+     * /admin/users and /admin/settings/advanced)
      */
     exclude_items?: Array<string>;
 
     /**
-     * List of path patterns to include. Supports wildcards (e.g., _/blog/_.html,
-     * /docs/\*_, _\blog\*.html)
+     * List of path patterns to include. Uses micromatch glob syntax: \* matches within
+     * a path segment, ** matches across path segments (e.g., /blog/** matches
+     * /blog/post and /blog/2024/post)
      */
     include_items?: Array<string>;
 
@@ -2142,22 +3153,65 @@ export namespace InstanceUpdateParams {
 
   export namespace SourceParams {
     export interface WebCrawler {
+      crawl_options?: WebCrawler.CrawlOptions;
+
       parse_options?: WebCrawler.ParseOptions;
 
-      parse_type?: 'sitemap' | 'feed-rss';
+      parse_type?: 'sitemap' | 'feed-rss' | 'crawl';
 
       store_options?: WebCrawler.StoreOptions;
     }
 
     export namespace WebCrawler {
+      export interface CrawlOptions {
+        depth?: number;
+
+        include_external_links?: boolean;
+
+        include_subdomains?: boolean;
+
+        max_age?: number;
+
+        source?: 'all' | 'sitemaps' | 'links';
+      }
+
       export interface ParseOptions {
+        /**
+         * List of path-to-selector mappings for extracting specific content from crawled
+         * pages. Each entry pairs a URL glob pattern with a CSS selector. The first
+         * matching path wins. Only the matched HTML fragment is stored and indexed.
+         */
+        content_selector?: Array<ParseOptions.ContentSelector>;
+
         include_headers?: { [key: string]: string };
 
         include_images?: boolean;
 
+        /**
+         * List of specific sitemap URLs to use for crawling. Only valid when parse_type is
+         * 'sitemap'.
+         */
+        specific_sitemaps?: Array<string>;
+
         use_browser_rendering?: boolean;
       }
 
+      export namespace ParseOptions {
+        export interface ContentSelector {
+          /**
+           * Glob pattern to match against the page URL path. Uses standard glob syntax: \*
+           * matches within a segment, \*\* crosses directories.
+           */
+          path: string;
+
+          /**
+           * CSS selector to extract content from pages matching the path pattern. Supports
+           * standard CSS selectors including class, ID, element, and attribute selectors.
+           */
+          selector: string;
+        }
+      }
+
       export interface StoreOptions {
         storage_id: string;
 
@@ -2171,12 +3225,27 @@ export namespace InstanceUpdateParams {
 
 export interface InstanceListParams extends V4PagePaginationArrayParams {
   /**
-   * Path param:
+   * Path param
    */
   account_id: string;
 
   /**
-   * Query param: Search by id
+   * Query param: Filter by namespace.
+   */
+  namespace?: string;
+
+  /**
+   * Query param: Field to order results by.
+   */
+  order_by?: 'created_at';
+
+  /**
+   * Query param: Order direction.
+   */
+  order_by_direction?: 'asc' | 'desc';
+
+  /**
+   * Query param: Filter instances whose id contains this string (case-insensitive).
    */
   search?: string;
 }
@@ -2185,17 +3254,345 @@ export interface InstanceDeleteParams {
   account_id: string;
 }
 
+export interface InstanceChatCompletionsParams {
+  /**
+   * Path param
+   */
+  account_id: string;
+
+  /**
+   * Body param
+   */
+  messages: Array<InstanceChatCompletionsParams.Message>;
+
+  /**
+   * Body param
+   */
+  ai_search_options?: InstanceChatCompletionsParams.AISearchOptions;
+
+  /**
+   * Body param
+   */
+  model?:
+    | '@cf/meta/llama-3.3-70b-instruct-fp8-fast'
+    | '@cf/zai-org/glm-4.7-flash'
+    | '@cf/meta/llama-3.1-8b-instruct-fast'
+    | '@cf/meta/llama-3.1-8b-instruct-fp8'
+    | '@cf/meta/llama-4-scout-17b-16e-instruct'
+    | '@cf/qwen/qwen3-30b-a3b-fp8'
+    | '@cf/deepseek-ai/deepseek-r1-distill-qwen-32b'
+    | '@cf/moonshotai/kimi-k2-instruct'
+    | '@cf/google/gemma-3-12b-it'
+    | '@cf/google/gemma-4-26b-a4b-it'
+    | '@cf/moonshotai/kimi-k2.5'
+    | 'anthropic/claude-3-7-sonnet'
+    | 'anthropic/claude-sonnet-4'
+    | 'anthropic/claude-opus-4'
+    | 'anthropic/claude-3-5-haiku'
+    | 'cerebras/qwen-3-235b-a22b-instruct'
+    | 'cerebras/qwen-3-235b-a22b-thinking'
+    | 'cerebras/llama-3.3-70b'
+    | 'cerebras/llama-4-maverick-17b-128e-instruct'
+    | 'cerebras/llama-4-scout-17b-16e-instruct'
+    | 'cerebras/gpt-oss-120b'
+    | 'google-ai-studio/gemini-2.5-flash'
+    | 'google-ai-studio/gemini-2.5-pro'
+    | 'grok/grok-4'
+    | 'groq/llama-3.3-70b-versatile'
+    | 'groq/llama-3.1-8b-instant'
+    | 'openai/gpt-5'
+    | 'openai/gpt-5-mini'
+    | 'openai/gpt-5-nano'
+    | '';
+
+  /**
+   * Body param
+   */
+  stream?: boolean;
+
+  [k: string]: unknown;
+}
+
+export namespace InstanceChatCompletionsParams {
+  export interface Message {
+    content: string | null;
+
+    role: 'system' | 'developer' | 'user' | 'assistant' | 'tool';
+
+    [k: string]: unknown;
+  }
+
+  export interface AISearchOptions {
+    cache?: AISearchOptions.Cache;
+
+    query_rewrite?: AISearchOptions.QueryRewrite;
+
+    reranking?: AISearchOptions.Reranking;
+
+    retrieval?: AISearchOptions.Retrieval;
+  }
+
+  export namespace AISearchOptions {
+    export interface Cache {
+      cache_threshold?: 'super_strict_match' | 'close_enough' | 'flexible_friend' | 'anything_goes';
+
+      enabled?: boolean;
+    }
+
+    export interface QueryRewrite {
+      enabled?: boolean;
+
+      model?:
+        | '@cf/meta/llama-3.3-70b-instruct-fp8-fast'
+        | '@cf/zai-org/glm-4.7-flash'
+        | '@cf/meta/llama-3.1-8b-instruct-fast'
+        | '@cf/meta/llama-3.1-8b-instruct-fp8'
+        | '@cf/meta/llama-4-scout-17b-16e-instruct'
+        | '@cf/qwen/qwen3-30b-a3b-fp8'
+        | '@cf/deepseek-ai/deepseek-r1-distill-qwen-32b'
+        | '@cf/moonshotai/kimi-k2-instruct'
+        | '@cf/google/gemma-3-12b-it'
+        | '@cf/google/gemma-4-26b-a4b-it'
+        | '@cf/moonshotai/kimi-k2.5'
+        | 'anthropic/claude-3-7-sonnet'
+        | 'anthropic/claude-sonnet-4'
+        | 'anthropic/claude-opus-4'
+        | 'anthropic/claude-3-5-haiku'
+        | 'cerebras/qwen-3-235b-a22b-instruct'
+        | 'cerebras/qwen-3-235b-a22b-thinking'
+        | 'cerebras/llama-3.3-70b'
+        | 'cerebras/llama-4-maverick-17b-128e-instruct'
+        | 'cerebras/llama-4-scout-17b-16e-instruct'
+        | 'cerebras/gpt-oss-120b'
+        | 'google-ai-studio/gemini-2.5-flash'
+        | 'google-ai-studio/gemini-2.5-pro'
+        | 'grok/grok-4'
+        | 'groq/llama-3.3-70b-versatile'
+        | 'groq/llama-3.1-8b-instant'
+        | 'openai/gpt-5'
+        | 'openai/gpt-5-mini'
+        | 'openai/gpt-5-nano'
+        | '';
+
+      rewrite_prompt?: string;
+    }
+
+    export interface Reranking {
+      enabled?: boolean;
+
+      match_threshold?: number;
+
+      model?: '@cf/baai/bge-reranker-base' | '';
+    }
+
+    export interface Retrieval {
+      /**
+       * Metadata fields to boost search results by. Overrides the instance-level
+       * boost_by config. Direction defaults to 'asc' for numeric/datetime fields,
+       * 'exists' for text/boolean fields. Fields must match 'timestamp' or a defined
+       * custom_metadata field.
+       */
+      boost_by?: Array<Retrieval.BoostBy>;
+
+      context_expansion?: number;
+
+      filters?: { [key: string]: unknown };
+
+      fusion_method?: 'max' | 'rrf';
+
+      /**
+       * Controls which documents are candidates for BM25 scoring. 'and' restricts
+       * candidates to documents containing all query terms; 'or' includes any document
+       * containing at least one term, ranked by BM25 relevance. Defaults to 'and'.
+       */
+      keyword_match_mode?: 'and' | 'or';
+
+      match_threshold?: number;
+
+      max_num_results?: number;
+
+      retrieval_type?: 'vector' | 'keyword' | 'hybrid';
+
+      return_on_failure?: boolean;
+    }
+
+    export namespace Retrieval {
+      export interface BoostBy {
+        /**
+         * Metadata field name to boost by. Use 'timestamp' for document freshness, or any
+         * custom_metadata field. Numeric and datetime fields support asc/desc directions;
+         * text/boolean fields support exists/not_exists.
+         */
+        field: string;
+
+        /**
+         * Boost direction. 'desc' = higher values rank higher (e.g. newer timestamps).
+         * 'asc' = lower values rank higher. 'exists' = boost chunks that have the field.
+         * 'not_exists' = boost chunks that lack the field. Optional ��� defaults to 'asc'
+         * for numeric/datetime fields, 'exists' for text/boolean fields.
+         */
+        direction?: 'asc' | 'desc' | 'exists' | 'not_exists';
+      }
+    }
+  }
+}
+
 export interface InstanceReadParams {
   account_id: string;
 }
 
+export interface InstanceSearchParams {
+  /**
+   * Path param
+   */
+  account_id: string;
+
+  /**
+   * Body param
+   */
+  ai_search_options?: InstanceSearchParams.AISearchOptions;
+
+  /**
+   * Body param
+   */
+  messages?: Array<InstanceSearchParams.Message>;
+
+  /**
+   * Body param: A simple text query string. Alternative to 'messages' — provide
+   * either this or 'messages', not both.
+   */
+  query?: string;
+}
+
+export namespace InstanceSearchParams {
+  export interface AISearchOptions {
+    cache?: AISearchOptions.Cache;
+
+    query_rewrite?: AISearchOptions.QueryRewrite;
+
+    reranking?: AISearchOptions.Reranking;
+
+    retrieval?: AISearchOptions.Retrieval;
+  }
+
+  export namespace AISearchOptions {
+    export interface Cache {
+      cache_threshold?: 'super_strict_match' | 'close_enough' | 'flexible_friend' | 'anything_goes';
+
+      enabled?: boolean;
+    }
+
+    export interface QueryRewrite {
+      enabled?: boolean;
+
+      model?:
+        | '@cf/meta/llama-3.3-70b-instruct-fp8-fast'
+        | '@cf/zai-org/glm-4.7-flash'
+        | '@cf/meta/llama-3.1-8b-instruct-fast'
+        | '@cf/meta/llama-3.1-8b-instruct-fp8'
+        | '@cf/meta/llama-4-scout-17b-16e-instruct'
+        | '@cf/qwen/qwen3-30b-a3b-fp8'
+        | '@cf/deepseek-ai/deepseek-r1-distill-qwen-32b'
+        | '@cf/moonshotai/kimi-k2-instruct'
+        | '@cf/google/gemma-3-12b-it'
+        | '@cf/google/gemma-4-26b-a4b-it'
+        | '@cf/moonshotai/kimi-k2.5'
+        | 'anthropic/claude-3-7-sonnet'
+        | 'anthropic/claude-sonnet-4'
+        | 'anthropic/claude-opus-4'
+        | 'anthropic/claude-3-5-haiku'
+        | 'cerebras/qwen-3-235b-a22b-instruct'
+        | 'cerebras/qwen-3-235b-a22b-thinking'
+        | 'cerebras/llama-3.3-70b'
+        | 'cerebras/llama-4-maverick-17b-128e-instruct'
+        | 'cerebras/llama-4-scout-17b-16e-instruct'
+        | 'cerebras/gpt-oss-120b'
+        | 'google-ai-studio/gemini-2.5-flash'
+        | 'google-ai-studio/gemini-2.5-pro'
+        | 'grok/grok-4'
+        | 'groq/llama-3.3-70b-versatile'
+        | 'groq/llama-3.1-8b-instant'
+        | 'openai/gpt-5'
+        | 'openai/gpt-5-mini'
+        | 'openai/gpt-5-nano'
+        | '';
+
+      rewrite_prompt?: string;
+    }
+
+    export interface Reranking {
+      enabled?: boolean;
+
+      match_threshold?: number;
+
+      model?: '@cf/baai/bge-reranker-base' | '';
+    }
+
+    export interface Retrieval {
+      /**
+       * Metadata fields to boost search results by. Overrides the instance-level
+       * boost_by config. Direction defaults to 'asc' for numeric/datetime fields,
+       * 'exists' for text/boolean fields. Fields must match 'timestamp' or a defined
+       * custom_metadata field.
+       */
+      boost_by?: Array<Retrieval.BoostBy>;
+
+      context_expansion?: number;
+
+      filters?: { [key: string]: unknown };
+
+      fusion_method?: 'max' | 'rrf';
+
+      /**
+       * Controls which documents are candidates for BM25 scoring. 'and' restricts
+       * candidates to documents containing all query terms; 'or' includes any document
+       * containing at least one term, ranked by BM25 relevance. Defaults to 'and'.
+       */
+      keyword_match_mode?: 'and' | 'or';
+
+      match_threshold?: number;
+
+      max_num_results?: number;
+
+      retrieval_type?: 'vector' | 'keyword' | 'hybrid';
+
+      return_on_failure?: boolean;
+    }
+
+    export namespace Retrieval {
+      export interface BoostBy {
+        /**
+         * Metadata field name to boost by. Use 'timestamp' for document freshness, or any
+         * custom_metadata field. Numeric and datetime fields support asc/desc directions;
+         * text/boolean fields support exists/not_exists.
+         */
+        field: string;
+
+        /**
+         * Boost direction. 'desc' = higher values rank higher (e.g. newer timestamps).
+         * 'asc' = lower values rank higher. 'exists' = boost chunks that have the field.
+         * 'not_exists' = boost chunks that lack the field. Optional ��� defaults to 'asc'
+         * for numeric/datetime fields, 'exists' for text/boolean fields.
+         */
+        direction?: 'asc' | 'desc' | 'exists' | 'not_exists';
+      }
+    }
+  }
+
+  export interface Message {
+    content: string | null;
+
+    role: 'system' | 'developer' | 'user' | 'assistant' | 'tool';
+
+    [k: string]: unknown;
+  }
+}
+
 export interface InstanceStatsParams {
   account_id: string;
 }
 
 Instances.InstanceListResponsesV4PagePaginationArray = InstanceListResponsesV4PagePaginationArray;
-Instances.Items = Items;
-Instances.ItemListResponsesV4PagePaginationArray = ItemListResponsesV4PagePaginationArray;
 Instances.Jobs = Jobs;
 Instances.JobListResponsesV4PagePaginationArray = JobListResponsesV4PagePaginationArray;
 
@@ -2205,26 +3602,21 @@ export declare namespace Instances {
     type InstanceUpdateResponse as InstanceUpdateResponse,
     type InstanceListResponse as InstanceListResponse,
     type InstanceDeleteResponse as InstanceDeleteResponse,
+    type InstanceChatCompletionsResponse as InstanceChatCompletionsResponse,
     type InstanceReadResponse as InstanceReadResponse,
+    type InstanceSearchResponse as InstanceSearchResponse,
     type InstanceStatsResponse as InstanceStatsResponse,
     InstanceListResponsesV4PagePaginationArray as InstanceListResponsesV4PagePaginationArray,
     type InstanceCreateParams as InstanceCreateParams,
     type InstanceUpdateParams as InstanceUpdateParams,
     type InstanceListParams as InstanceListParams,
     type InstanceDeleteParams as InstanceDeleteParams,
+    type InstanceChatCompletionsParams as InstanceChatCompletionsParams,
     type InstanceReadParams as InstanceReadParams,
+    type InstanceSearchParams as InstanceSearchParams,
     type InstanceStatsParams as InstanceStatsParams,
   };
 
-  export {
-    Items as Items,
-    type ItemListResponse as ItemListResponse,
-    type ItemGetResponse as ItemGetResponse,
-    ItemListResponsesV4PagePaginationArray as ItemListResponsesV4PagePaginationArray,
-    type ItemListParams as ItemListParams,
-    type ItemGetParams as ItemGetParams,
-  };
-
   export {
     Jobs as Jobs,
     type JobCreateResponse as JobCreateResponse,