You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: _posts/2025-10-16-Introducing-Real-Time-Streaming-for-AI-Models-and-Agents-in-OpenSearch.md
+11-11Lines changed: 11 additions & 11 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -155,7 +155,7 @@ POST /_plugins/_ml/models/yFT0m5kB-SbOBOkMDNIa/_predict/stream
155
155
156
156
#### Sample response
157
157
158
-
The streaming format uses Server-Sent Events (SSE), with each chunk containing a portion of the model's response. Each data line represents a separate chunk transmitted in real-time as the model generates output:
158
+
The streaming format uses Server-Sent Events (SSE), with each chunk containing a portion of the model's response. Each data line represents a separate chunk transmitted in realtime as the model generates output:
*`content` - The text fragment generated in this chunk (for example, a word, or phrase)
179
-
*`is_last` - A Boolean flag indicating whether this is the final chunk (`false` for intermediate chunks, `true` for the last one)
178
+
*`content` -- The text fragment generated in this chunk (for example, a word, or phrase).
179
+
*`is_last` -- A Boolean flag indicating whether this is the final chunk (`false` for intermediate chunks, `true` for the last one).
180
180
181
181
### Step 2: Set up agent streaming
182
182
@@ -291,7 +291,7 @@ POST /_plugins/_ml/agents/37YmxZkBphfsuvK7qIj4/_execute/stream
291
291
292
292
#### Sample response
293
293
294
-
The streaming format uses SSE, with each chunk containing a portion of the agent's response. Each data line represents a separate chunk transmitted in real-time as the agent generates output.
294
+
The streaming format uses SSE, with each chunk containing a portion of the agent's response. Each data line represents a separate chunk transmitted in realtime as the agent generates output:
*`content` - The text or data fragment generated in this chunk (for example, a word or phrase).
331
-
*`is_last` - A Boolean flag indicating whether this is the final chunk (`false` for intermediate chunks, `true` for the last one).
332
-
*`memory_id` - A unique identifier for the conversation memory session.
333
-
*`parent_interaction_id` - An identifier linking related interactions in the conversation.
330
+
*`content` -- The text or data fragment generated in this chunk (for example, a word or phrase).
331
+
*`is_last` -- A Boolean flag indicating whether this is the final chunk (`false` for intermediate chunks, `true` for the last one).
332
+
*`memory_id` -- A unique identifier for the conversation memory session.
333
+
*`parent_interaction_id` -- An identifier linking related interactions in the conversation.
334
334
335
335
## Conclusion
336
336
@@ -340,6 +340,6 @@ Streaming capabilities in OpenSearch represent a significant step forward in del
340
340
341
341
## What's next?
342
342
343
-
* Explore the official documentation for [Predict Stream](https://docs.opensearch.org/latest/ml-commons-plugin/api/train-predict/predict-stream/) and [Execute Stream Agent](https://docs.opensearch.org/latest/ml-commons-plugin/api/agent-apis/execute-stream-agent/) API references
344
-
* Share your feedback on [OpenSearch forum](https://forum.opensearch.org/)
345
-
* Stay tuned for updates as streaming support expands in future releases
343
+
* Explore the [Predict Stream](https://docs.opensearch.org/latest/ml-commons-plugin/api/train-predict/predict-stream/) and [Execute Stream Agent](https://docs.opensearch.org/latest/ml-commons-plugin/api/agent-apis/execute-stream-agent/) API references.
344
+
* Share your feedback on the [OpenSearch forum](https://forum.opensearch.org/).
345
+
* Stay tuned for updates as streaming support expands in future releases.
0 commit comments