Letta Cloud alternatives
Choose when the agent IS a long-lived stateful runtime. For bolt-on memory use Mem0/Zep.
Overall 7.4 (curated_estimate, medium) · score asOf 2026-09-14 · last reviewed 2026-09-14. This is not an Elo and not a live price board. Provider · JSON
Jobs: agent-backend, rag-memory, agent-memory-kv, hosted-agent-memory, temporal-graph-memory, stateful-agent-runtime, episodic-long-term-memory
| Provider | Why listed | Overall | Δ vs subject | asOf |
|---|---|---|---|---|
| Mem0 Default hosted agent memory bolt-on. Model retrieval volume carefully; self-host OSS if caps hurt. | same_jtbd_pick, same_jtbd_runner · hosted-agent-memory, temporal-graph-memory, stateful-agent-runtime | 7.59high | +0.19 | 2026-09-14 |
| Supabase (pgvector) Depot default for most agent apps under ~25–50M vectors. Don't add Pinecone until Postgres actually hurts. | same_jtbd_pick, same_jtbd_runner · rag-memory, agent-memory-kv, hosted-agent-memory | 8.64high | +1.24 | 2026-09-14 |
| Zep Pick when agents need temporal/entity graph memory, not just vector snippets. Mem0 for simpler recall. | same_jtbd_runner, same_jtbd_pick · hosted-agent-memory, temporal-graph-memory, episodic-long-term-memory | 7.44high | +0.04 | 2026-09-14 |
| Redis Use as agent hot memory/cache, not as your only 100M-vector archive. | same_jtbd_runner, same_jtbd_pick · rag-memory, agent-memory-kv | 8.07high | +0.67 | 2026-09-14 |
| Modal Depot darling for Python agent backends and bursty GPU jobs. If you think in functions not VMs, start here. | same_jtbd_pick · agent-backend | 8.44high | +1.04 | 2026-10-19 |
| Upstash Redis Default serverless-edge-redis. Use classic Redis Cloud when you need always-on fat memory / Redis modules… | same_jtbd_runner · rag-memory, agent-memory-kv | 8.49high | +1.09 | 2026-09-14 |
| Cognee Graph memory when documents→ontology matter. Zep for temporal bipartite graphs; Mem0 for simplest hosted… | same_jtbd_runner · hosted-agent-memory, temporal-graph-memory | 8.09high | +0.69 | 2026-09-14 |
| Memobase Depot pick for portable/user-profile memory (MCP across Claude/ChatGPT/agents). Mem0 remains general… | same_jtbd_runner · hosted-agent-memory, episodic-long-term-memory | 7.94high | +0.54 | 2026-09-14 |
| Supermemory Managed context engine when you want memory without running a vector DB. Mem0 for simpler fact API; Cognee… | same_jtbd_runner · hosted-agent-memory, episodic-long-term-memory | 7.88high | +0.48 | 2026-09-14 |
| Qdrant Depot pick for serious vector workloads with filters. Self-host or Cloud — either way you can leave. | same_jtbd_runner · rag-memory | 8.59high | +1.19 | 2026-10-19 |
| Cloudflare Workers Best edge runtime for lightweight agent gateways and tools. Heavy LLM loops still belong on Modal/Fly/GPU… | same_jtbd_runner · agent-memory-kv | 8.53high | +1.13 | 2026-09-14 |
| Fly.io Excellent for globally distributed agent sidecars and Machines-based sandboxes. Prefer Railway/Render if you… | same_jtbd_runner · agent-backend | 8.28high | +0.88 | 2026-09-14 |
Prices are not ranked on this page. For a job, use the sourced quote (separate input and output rows, Depot Pick unchanged): agent-backend, rag-memory, agent-memory-kv, hosted-agent-memory.
Alternatives are other catalog providers that share a JTBD as pick or runner-up. If fewer than three exist, same-category active providers fill the list and are labeled same_category. Overall deltas are catalog scores. kind=curated_estimate is not a lab measurement. Pricing rows are sourced or explicitly null. Do not parse pricingNotes. A single normalized USD figure is set only for a provider with exactly one token row.