{% extends "base.html" %} {% block content %}
Memory Test Lab

Ask a local model with your 8mem context attached.

This is a private local sandbox for checking whether your saved preferences and corrections actually change the answer before you use them in Telegram or another assistant.

See what AI knows Review memory library
Prompt

Ask with memory

Ollama

Model: {{ model }}
Ollama URL: {{ base_url }}

Answer

Memory-aware answer

Local
{% if error %}

If Ollama is not installed, install it later and rerun 8mem setup. Telegram memory still works without this local model test.

{% elif answer %}
{{ answer }}
{% else %}
No answer yet. Ask a question to test memory-aware prompting.
{% endif %}
What this proves

Your memory is portable beyond one chat surface

Telegram is the first launch channel. This local test proves the same memory can shape answers wherever a model or agent chooses to read 8mem context.

{% endblock %}