Ollama's Responses API accepts previous_response_id, returns 200, and forgets the whole conversation
TL;DR : send previous_response_id to Ollama's /v1/responses and it is accepted, the reply is HTTP 200 with "status": "completed" and "error": null , and the model has never seen the previous turn. A request that carries previous_response_id costs the same number of input tokens as a request that carries no history — 41 and 41 on this box — and produces the same kind of answer. Sending the full…
Ollama's Responses API accepts a previous_response_id in requests, returns a 200 status, but forgets the entire conversation context. The model treats each request as a new conversation, regardless of the previous_response_id sent. This is due to the request struct lacking a previous_response_id field, which gets discarded during JSON decoding.
The server's response includes a null previous_response_id field, indicating the feature is not supported. To resolve this, clients should send the full conversation history with each request to ensure the model has the necessary context.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
