@tanstack/ai 0.26.1 → 0.27.0

package/src/types.ts

@@ -812,41 +812,6 @@ export interface TextOptions<
    */
   systemPrompts?: Array<SystemPrompt>
   agentLoopStrategy?: AgentLoopStrategy
-  /**
-   * Controls the randomness of the output.
-   * Higher values (e.g., 0.8) make output more random, lower values (e.g., 0.2) make it more focused and deterministic.
-   * Range: [0.0, 2.0]
-   *
-   * Note: Generally recommended to use either temperature or topP, but not both.
-   *
-   * Provider usage:
-   * - OpenAI: `temperature` (number) - in text.top_p field
-   * - Anthropic: `temperature` (number) - ranges from 0.0 to 1.0, default 1.0
-   * - Gemini: `generationConfig.temperature` (number) - ranges from 0.0 to 2.0
-   */
-  temperature?: number
-  /**
-   * Nucleus sampling parameter. An alternative to temperature sampling.
-   * The model considers the results of tokens with topP probability mass.
-   * For example, 0.1 means only tokens comprising the top 10% probability mass are considered.
-   *
-   * Note: Generally recommended to use either temperature or topP, but not both.
-   *
-   * Provider usage:
-   * - OpenAI: `text.top_p` (number)
-   * - Anthropic: `top_p` (number | null)
-   * - Gemini: `generationConfig.topP` (number)
-   */
-  topP?: number
-  /**
-   * The maximum number of tokens to generate in the response.
-   *
-   * Provider usage:
-   * - OpenAI: `max_output_tokens` (number) - includes visible output and reasoning tokens
-   * - Anthropic: `max_tokens` (number, required) - range x >= 1
-   * - Gemini: `generationConfig.maxOutputTokens` (number)
-   */
-  maxTokens?: number
   /**
    * Additional metadata to attach to the request.
    * Can be used for tracking, debugging, or passing custom information.

package/src/utilities/sampling-keys.ts

@@ -0,0 +1,28 @@
+/**
+ * Single source of truth for the provider-native key spellings that cap output
+ * tokens. Sampling options live in opaque, provider-native `modelOptions`, and
+ * every provider spells the token cap differently. Two call sites must agree on
+ * this set or they silently drift:
+ *
+ * - `activities/summarize/chat-stream-summarize.ts` — detects a caller-supplied
+ *   token limit so the summarize default never overrides it.
+ * - `middlewares/otel.ts` — picks the first numeric spelling to populate the
+ *   `gen_ai.request.max_tokens` attribute across providers.
+ *
+ * Keep this list in lockstep with `MAX_TOKENS_KEY_BY_ADAPTER` (the adapter →
+ * native-key map) in the summarize wrapper.
+ */
+export const MAX_TOKENS_KEYS = [
+  'max_output_tokens', // OpenAI (Responses)
+  'max_tokens', // Anthropic / Grok
+  'max_completion_tokens', // Groq
+  'maxOutputTokens', // Gemini
+  'maxCompletionTokens', // OpenRouter
+  'maxTokens', // generic / migration leftover (no adapter reads it)
+] as const
+
+/**
+ * Ollama nests sampling under `options`; its token cap is `options.num_predict`
+ * rather than a flat key.
+ */
+export const NESTED_MAX_TOKENS_KEY = 'num_predict' as const