Different products. Different use cases.
This table links to supermemory's own documentation for each detail. Follow the source links for their latest features.
Swipe horizontally to compare all three columns →
| Dimension | ContextStream | supermemory |
|---|---|---|
| Product center | Project memory for AI coding agents: decisions, lessons, plans, handoffs, and code search in one MCP-accessible layer. | A general memory layer and second brain for AI agents and applications. supermemory docs ↗ |
| MCP access and editors | MCP server connects Cursor, Claude Code, Cline, and Continue to project memory. Code search and structured tools included. | Provides a memory API. Integration with specific coding assistants depends on your implementation. API reference ↗ |
| What is remembered | Decisions, lessons, plans, handoffs, code search results, dependency graphs, and structured project knowledge. | General memories, notes, and content added via the API or UI. Memory model ↗ |
| Code search and graph | Built-in code search with 99.38% Recall@10. Dependency, call, dataflow, and type graphs sit alongside memory. | Not a code-search product. Memory retrieval focuses on general content. Search docs ↗ |
| Team and project scope | Workspaces, projects, team roles, and project-level access controls. Memory scoped to the right project and team members. | Users can organize memories and control access through spaces. Spaces docs ↗ |
| LongMemEval-S accuracy | 90.0% on LongMemEval-S (ContextStream-published). View benchmarks → | 85.4% on LongMemEval-S (vendor-published by supermemory). supermemory benchmarks ↗ |
Choose for the product center that fits.
You want a memory API or second brain.
supermemory is the natural fit when you need a general-purpose memory layer, a second brain for personal knowledge, or an API you can integrate into your own agent or application.
Your next Cursor session should already know the decision.
ContextStream is built for coding projects. Code search, decisions, lessons, plans, handoffs, dependency graphs, and continuity across Cursor and Claude Code sessions should arrive as one layer.
Test with Starter. Scale with Pro or Elite.
Pro starts at $19/month with 25,000 Context Credits, persistent memory, and full product access. Elite starts at $49/month for deeper graph context and 100,000 monthly credits. Starter includes 3,000 credits and 5-day memory retention.
Read the method, not just the score.
ContextStream reports 90.0% accuracy on LongMemEval-S and 99.38% Recall@10 in its published code-search evaluation. These are ContextStream measurements on ContextStream tasks.
Evaluate ContextStream with your project.
Connect Cursor or Claude Code, index one repository, and test whether the next session retrieves a real decision and the code it affects.