agent-lattice 0.17.0 → 0.17.1

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/README.md CHANGED
@@ -140,10 +140,16 @@ const result = await agent.prompt("Solve this carefully.", {
140
140
 
141
141
  Use `{ type: "adaptive" }` for models that support adaptive thinking. Use
142
142
  `{ type: "enabled", budgetTokens }` for models that require a fixed budget, or
143
- `{ type: "disabled" }` to omit thinking from the provider request. A fixed
143
+ `{ type: "disabled" }` to turn thinking off. A fixed
144
144
  budget is capped at `maxTokens - 1` to satisfy the Anthropic API constraint.
145
145
  When omitted, the SDK does not send a thinking configuration.
146
146
 
147
+ `{ type: "disabled" }` is sent to the provider explicitly as
148
+ `thinking: { "type": "disabled" }` rather than omitted, because some
149
+ Anthropic-compatible providers (for example DeepSeek's
150
+ `https://api.deepseek.com/anthropic` endpoint) default thinking to on —
151
+ omitting the field would leave it enabled.
152
+
147
153
  For Kimi K3 through an Anthropic-compatible endpoint or gateway, use
148
154
  `reasoningEffort` to send the provider's top-level `reasoning_effort` parameter:
149
155