For developers building with AI memory.
Everything in Free, plus
- Auto top-up with configurable purchase limits
- 3 team seats, unlimited end users
- Priority support
Pay for what you use.
Credits included on every plan.
Every plan comes with a monthly balance of credits. Calls draw from it at the same rates whichever plan you are on, and only unique content is ever billed.
Tried almost everything — structured memory files, qmd etc. The only thing that works reliably is supermemory.
Not facing any memory issues after setting it up.
Harshil MathurFounder, Razorpay100B+tokens a month187msmedian recall
Plans
For developers who need more headroom than Pro.
Everything in Pro, plus
- 6.5× the credits of Pro
- Gmail connector
For teams and production workloads.
Everything in Max, plus
- Advanced document extraction for PDFs
- S3 and web crawler connectors
- Unlimited team seats, per-tag access
- Dedicated support
- SOC 2 · HIPAA BAA · self-hosted option
Enterprise
Custom deployments with dedicated engineering.
For organizations with committed spend, custom deployments and their own security requirements. Unlimited usage, metered your way.
Security and compliance
- Air-gapped self-hosting or a dedicated managed instance
- SOC 2 · HIPAA · GDPR
- Custom contracts and DPA
- SSO and custom integrations
- Enterprise MCP
Scale and performance
- Unlimited usage on committed-spend pricing
- Advanced document extraction for PDFs
- Custom metering and billing
- Custom rate limits and throughput
- Dedicated infrastructure
Support
- Dedicated account manager
- Forward-deployed engineer
- 1:1 onboarding and integration
- Uptime SLA
- Priority Slack channel
Rate card
Billed in SM tokens, the unique tokens we actually ingest. Repeats and unchanged content are free, and the rates are the same on every plan.
- MemoryA memory graph per user: profiles and fact hierarchies that agents learn from in real time. Powered by our own model.
Plain text$5/ 1M SM tokens
Rich content$10/ 1M SM tokens
- SuperRAGMultimodal extraction, contextual chunking and retrieval. No embeddings or vectors to manage. Also available as a filesystem. smfs.ai
Text$1/ 1M SM tokens
Rich: images, PDFs, audio, video$2/ 1M SM tokens
- Search and traversalSemantic search and graph traversal across stored content, in one call. 187ms median server time, built for agent loops.
Per query$5/ 1M queries
- OperationsRe-ranking, aggregation, query rewriting and other building blocks for richer queries.
Per operation$100/ 1M operations
Questions
- How does the credit balance work?
- Every plan comes with a monthly dollar balance. Each API call, whether storing memories, searching or indexing, draws from that balance at the rates above. Top-ups add to it whenever you like.
- What is an SM token?
- SM tokens measure billable content processed during ingestion. Supported updates to the same document bill the net-new token delta; search and other operations are metered separately.
- What happens if I store the same content twice?
- For supported updates using the same customId within the same organization, unchanged content is not billed again on the token meter. New content and separately metered operations can still incur charges. A new document identity or a full replacement can change how ingestion is billed.
- Do unused credits roll over?
- Subscription credits reset monthly. Credits you buy as a top-up never expire; they stay in your balance until you use them.
- What happens when I run out?
- Usage requires an available credit balance. You can purchase additional credits in the console. Paid plans also offer automatic top-ups with configurable purchase limits. These limits control automatic credit purchases; they are not a cap on your total invoice.
- Which plans include advanced document extraction?
- Scale and Enterprise include advanced PDF extraction: page-by-page OCR with descriptions of figures and diagrams alongside the text. It is selected automatically based on your organization's plan, with standard extraction as a fallback if needed. Free, Pro and Max continue to support standard PDF extraction, including OCR for scans. This plan distinction applies to PDFs, not every supported file type.
- Can I self-host?
- Self-hosted deployments are available on Scale and Enterprise. Enterprise also supports fully air-gapped deployments, where LLM inference may be the only outbound call, or none at all.
- Do you offer startup or research credits?
- Yes. Qualifying early-stage startups and academic research teams get the Scale plan free for three months, including its included usage, plus dedicated support.