Appearance
Memory System
The memory system enables persistent, contextual conversations by storing and retrieving relevant information about your interactions. Unlike traditional chat applications that forget everything between sessions, Iris builds a growing understanding of you over time.
Two-Tier Recall
Iris uses a two-tier system to balance completeness with efficiency:
Tier 1: Important Memories
High-importance memories (≥0.80) always included in context, regardless of conversation topic. These are facts Iris should always know - your name, key preferences, important goals.
Tier 2: Semantic Search
Contextually relevant memories retrieved based on what you're currently discussing. The system generates search queries from recent conversation, finds semantically similar memories, and ranks them by relevance.
Memory Types
| Type | Description |
|---|---|
fact | Objective information about the user |
preference | Likes, dislikes, and preferences |
goal | Things the user wants to achieve |
event | Past or upcoming events |
skill | Capabilities and expertise |
relationship | People and relationships |
habit | Regular behaviors and routines |
context | Situational information |
Scoring
Memories are ranked by a composite score combining:
- Semantic similarity (45%) - How relevant to the current conversation
- Importance (30%) - The memory's assigned importance (0.0-1.0)
- Recency (15%) - Decays linearly over 90 days
- Frequency (10%) - How often the memory has been accessed
Certain types get score bonuses: relationships (+0.10), preferences and goals (+0.05), and recent events (+0.10).
Configuration
| Setting | Default | Description |
|---|---|---|
memory.tier1_max | 5 | Important memories always injected |
memory.tier2_max | 7 | Semantic search results |
memory.importance_threshold | 0.80 | Minimum importance for Tier 1 |
memory.similarity_threshold | 0.38 | Minimum similarity for Tier 2 |
TIP
If memories seem stale or irrelevant, adjust the scoring weights in config. Increase recency to prioritize recent memories, or semantic to prioritize relevance.
How Memories Are Created
- Automatic Extraction - Background job analyzes conversations and extracts memorable facts
- Consolidation - Similar memories are merged nightly, with generation tracking to enable organic memory evolution
- Manual Tools - Iris can store, update, or delete memories directly during conversation
Viewing Memories
The Memories page in the UI shows all your memories with filtering and search. Consolidated memories display their generation level (e.g., "Merged Gen 2") to show how many times they've been consolidated. The Insights page includes consolidation statistics in the Memory Insights panel.