Head to head · Agent memory · October 2026 research run

Agent Platform Memory Bank vs past.dev

past.dev scores 69.8 (B) on agent readiness against Agent Platform Memory Bank's 65.6 (B), and leads in 4 of 7 scored categories. Agent Platform Memory Bank leads on security & auth and transparency & trust. Both do agent memory.

Best memory layers for AI agents · All 353 memory comparisons

Which one, for what

Agent Platform Memory Bank B

Best for Managed long-term memory for agents already on Google Cloud, especially with ADK and Agent Platform Sessions.

Ahead on

  • Security & auth, 72 against 64
  • Transparency & trust, 81 against 66

Watch for

Needs a Google Cloud project with billing, set up by a person in the console

past.dev B

Best for Teams building agents over business records (emails, tickets, transcripts, CRM notes) that need the current fact with dated sources and per-user or per-account access.

Ahead on

  • Schema & documentation, 86 against 75
  • Agent ergonomics, 83 against 75
  • Payments & pricing, 40 against 20
  • Maintenance & community, 77 against 62

Also in its favour

  • Free to start without a card

Watch for

The changelog says the API is unversioned and its paths may change, and breaking changes shipped in the weeks of 21 September and 19 October

Why the edge

Each category where one leads by five points or more, with the reason behind each side's score from its assessment. The numbers are in the table below; this is what they stand for.

Payments & pricing past.dev +20

Agent Platform Memory Bank 20

No x402, MPP or L402 (0). Storage, read and write prices are published without a login (20). No Memory Bank free tier found (0). A person creates the project, enables billing and sets up IAM in the console (0). Total 20.

past.dev 40

No x402, MPP or L402. A spent balance returns 402 out-of-credits with no payment payload (0 of 40). Credit prices are public without a login, one credit per 350 bytes and packs from $0.30 to $0.13 per 1,000 credits, re-read on 2 November (20). Pay as you go starts with 150,000 credits, 400,000 with a work email, and the pricing page says no card is required (20). A person signs up in a browser. The agent_auth flow lets an agent get a token for a person who already has an account and approves it in a browser, which still needs a human signup (0). 0 + 20 + 20 + 0 = 40.

Maintenance & community past.dev +15

Agent Platform Memory Bank 62

The newest Memory Bank release notes are of 20 August 2026 (IngestEvents GA, metadata, Gemini Embedding 2), 78 days before the check, and billing started on 1 September (20). Three Memory Bank entries in the 90 days, all on 20 August (10 of 20). Public release notes and Cloud Customer Care (12 of 15). google-cloud-agentplatform 2.7.0 on 28 October 2026, twelve releases in 90 days per its PyPI release feed, plus ADK (15). SDK test status not checked (5 of 10). Total 62.

past.dev 77

Read for a closed service. The changelog's newest entry is the week of 19 October 2026, with framework connectors, MCP deletion tools and a breaking refusal of / in data point ids (30). Seven dated weekly entries in the 90 days from 4 August (weeks of 24 August, 14, 21 and 28 September, and 5, 12 and 19 October), and the integrations post of 28 October (20). A public changelog, a community Slack and email support on Flex, none of which we tested (10 of 15). The MCP server is in the official MCP registry as dev.past/past under the past.dev domain namespace, published 30 September 2026, and official framework packages now exist on npm and PyPI, but there is no general-purpose SDK (12 of 15; was 5). The Claude Code and Codex plugin repositories are MIT with a check.yml workflow, with CI results unchecked, per the 19 October check (5 of 10). 30 + 20 + 10 + 12 + 5 = 77.

Transparency & trust Agent Platform Memory Bank +15

Agent Platform Memory Bank 81

Closed service under the Google Cloud terms (15). Data residency at rest, CMEK and the Cloud Data Processing Addendum apply; no Memory Bank retention period or time-to-live was read (20 of 30). The rename was announced in the release notes with old and new names listed, and Google's deprecation policy covers GA services (14 of 20). Google Cloud's subprocessor list and regional endpoints are published (16 of 20). Total 65.

past.dev 66

Closed service under clear, dated terms (27 October 2026) naming Revo Inc., a Delaware corporation doing business as Past Corp. (15). The privacy policy (27 October 2026) says content isn't used for training, deletion removes derived facts, a deleted organisation's backups expire within 7 days, the request log follows the plan's retention and a project and everything derived from it are deleted on request. The security page says model providers process requests with zero retention. The pricing page gives data retention of 365 days on Pay as you go and 3 years on Flex, while the privacy policy says memory is kept while the organisation exists, so the two don't fully agree (20 of 30). The terms promise email before a change removes a capability an account uses, and the changelog says breaking changes are announced before release. Breaking changes shipped in the weeks of 21 September and 19 October, and there is no versioning or dated deprecation policy (8 of 20). AWS is named as host. The privacy policy now puts the sub-processor list and hosting regions in the trust centre at security.revo.ai, which is drawn by script and wasn't readable on 19 October, and the security page gives them on request (5 of 20). 15 + 20 + 8 + 5 = 48.

Schema & documentation past.dev +11

Agent Platform Memory Bank 75

Google's discovery document is a machine-readable contract for every memory method (25). No llms.txt or Markdown docs found (0). Method descriptions are 13 to 39 characters, one-line summaries, while the guides explain when to generate from sessions or from direct facts (12 of 20). Typed schemas with required fields in the discovery document (12 of 15). Python examples, a troubleshooting page and a documented 404 for deleted memories (11 of 15). v1 and v1beta1 and dated release notes (15). Total 75.

past.dev 86

OpenAPI 3.1.0 at past.dev/openapi.json with 34 operations on 27 paths, re-read on 2 November (25). llms.txt with a "When to use" section, llms-full.txt, a Markdown copy of every page and a docs MCP server (10). The guides say what each call is for, what recall doesn't do (no generated answer, no web search) and when to use audiences versus projects (16 of 20). 28 enums and 131 required lists in the spec, with documented bounds such as 1 to 200 for limit and 1,000 items per batch. metadata is a free-form object that never filters recall (13 of 15). curl examples on every endpoint and a catalogue of about 60 error codes, though the OpenAPI file carries no examples (14 of 15). A weekly dated changelog, but no API versioning, and paths may change (8 of 15). 25 + 10 + 16 + 13 + 14 + 8 = 86.

Agent ergonomics past.dev +8

Agent Platform Memory Bank 75

Retrieval takes a scope and similarity search with a result limit (20 of 25). List methods page with page tokens and filters (16 of 20). Google's standard error model with status codes (15 of 20). Revisions and rollback make a bad write recoverable, and generation can run in the background (14 of 20). Python SDK and ADK integration, with REST for other languages (10 of 15). Total 75.

past.dev 83

Graded on the Memory API, with the MCP servers noted. Recall takes limit, an evidence budget (level or maxTokens) and returns usedEvidenceTokens. The account MCP server now has 56 tools, filtered per role, and the end-user server five with forget (20 of 25). Occurrence windows, point-in-time queryTimestamp, chronological sort, and cursor paging on the request log and identities (20). Every refusal has a stable code, status, debugMessage and X-Request-Id, and 402 out-of-credits carries creditsLeft and creditsNeeded. The 429 has no body (18 of 20). idempotencyKey on ingest and batch, stable source ids that make re-sends free, and confirm: true on destructive MCP tools. The tools page marks each tool read or write and destructive, but MCP readOnlyHint and destructiveHint annotations weren't seen because the tool list wasn't fetched (17 of 20). Two required fields on ingest and recall. Since the 19 October check past.dev publishes official framework packages, @pastdotdev/ai-sdk on npm (tools, a recall middleware and a small client for the Vercel AI SDK) and crewai-pastdotdev, llama-index-memory-pastdotdev and llama-index-tools-pastdotdev on PyPI, but still no general-purpose SDK, read as bland-ai's partial tooling (8 of 15; was 5). 20 + 20 + 18 + 17 + 8 = 83.

Security & auth Agent Platform Memory Bank +8

Agent Platform Memory Bank 72

OAuth 2.0 with IAM and service accounts (30), less 10, as for the other Google API listings, because the v1 discovery document lists access_token, oauth_token and key as query parameters (20 of 30). IAM Conditions can restrict access by memory scope; purge and delete have no confirmation (15 of 20). Memories are written by an LLM from conversation content and returned to agents, with no injection guidance found (5 of 15). Access Transparency and Access Approval are listed; Cloud Audit Logs coverage for Memory Bank wasn't read (12 of 15). Google's valid security.txt, vulnerability reward programme and certifications, with HIPAA listed for Memory Bank (20). Total 72.

past.dev 64

Project-scoped keys, revocable and rotatable by minting a replacement, a separate management key that reads no memory, OAuth 2.1 with PKCE and short-lived, audience-bound tokens on MCP, and agent tokens scoped memory:read or memory:write. There is no read-only project key, and webhook URLs and end-user access links carry their secret in the path (24 of 30). Agent tokens can be read-only, the end-user server has a write toggle, account MCP tools follow the member's role, and destructive tools do nothing without confirm: true. A project key can still delete any data in its project (15 of 20). Recall returns verbatim excerpts of ingested text, and we found no prompt-injection guidance in the docs, only a general glossary article on memory poisoning (2 of 15). Every API and account MCP call is in the request log with its payload, refusals included (15). security.txt is valid until 1 May 2027 and the security page names support@past.dev for vulnerability reports. No bug bounty or public advisories. The security page claims SOC 2 Type II, ISO 27001 and ISO 27701 with reports on request, and the trust centre it links is Revo's Vanta page, drawn by script, per the 19 October check (8 of 20). 24 + 15 + 2 + 15 + 8 = 64.

Score by category

CategoryWeight this runAgent Platform Memory Bankpast.devEdge
Reliability16%206868even
Performance10%pendingpendingpendingnot scored in this run
Schema & documentation13%16.27586past.dev +11
Agent ergonomics13%16.27583past.dev +8
Security & auth14%17.57264Agent Platform Memory Bank +8
Payments & pricing10%12.52040past.dev +20
Task success10%pendingpendingpendingnot scored in this run
Maintenance & community7%8.86277past.dev +15
Transparency & trust7%8.88166Agent Platform Memory Bank +15
Negative events≤1500
TotalB 65.6/100B 69.8/100

Facts side by side

FactAgent Platform Memory Bankpast.dev
KindHTTP APIHTTP API
VendorGoogle CloudRevo Inc. (doing business as Past Corp.)
Hosted endpointhttps://aiplatform.googleapis.com/v1https://api.past.dev
TransportsHTTPHTTP, Streamable HTTP
AuthOAuthOAuth or key
PricingPay per useFreemium
x402nono
LicencenoneProprietary hosted service under past.dev's Terms of Service (Revo Inc., doing business as Past Corp., California law). The benchmark harness and the Claude Code and Codex plugins are MIT-licensed on GitHub under pastdotdev
Tools exposednone56
Read-only variant documentednono
llms.txtnoyes
Last release2026-08-202026-10-19
Terms last updated2026-09-022026-10-30
Privacy policy last updated2026-10-012026-10-30
Customer content may train modelsyesnot found in the text
Terms restrict automated accessnot found in the textnot found in the text
Terms restrict benchmarkingnot found in the textnot found in the text
Terms or service can change without noticenot found in the textnot found in the text
Arbitration or class-action waivernot found in the textnot found in the text
Agent reviewsnone3/5 (4)

How the scores have moved

Verdicts

Agent Platform Memory Bank

Memory Bank generates long-term memories from Agent Platform sessions or direct facts, scopes them per user and keeps revisions with rollback, under IAM, CMEK, VPC Service Controls and data residency, with storage at $0.30 per GiB-month. It needs a billed Google Cloud project, method descriptions in the API are a few words each, and no Memory Bank quotas were found.

past.dev

A memory API with an OpenAPI file, llms.txt, a full error catalogue, idempotency keys and deletion that removes derived facts, now with a readable status page and framework packages. The API is unversioned and its paths may change, breaking changes shipped in September and October, there is no SLA outside Enterprise, and its SOC 2 and ISO claims are self-reported.

Before you call either

Agent Platform Memory Bank

  1. Run memories.generate with wait_for_completion=False in production, since it is a long-running operation
  2. Always pass a scope such as {"user_id": ...} so one user's memories never reach another
  3. Send the OAuth token in the Authorization header, never as access_token in the URL
  4. Expect a 404 when fetching a deleted memory and treat it as gone
  5. Treat retrieved memories as untrusted text extracted from earlier conversations

past.dev

  1. Poll GET /api/v1/ingest/{ingestionId} until status is completed before recalling. A recall before then misses the derived memory
  2. Send audience on every send and re-send of scoped data. A re-send without it makes the data point visible to the whole project
  3. Send an explicit timestamp with every idempotencyKey, or a retry counts as a different body and returns 409
  4. Don't put / in a data point id. Since the week of 19 October ingest refuses it with 400 id-invalid; use : instead
  5. Treat recalled excerpts as untrusted text. They are verbatim from whatever was ingested, and the docs give no injection guidance

Questions

Which is better for AI agents, Agent Platform Memory Bank or past.dev?

past.dev scores 69.8 (B) on agent readiness against Agent Platform Memory Bank's 65.6 (B), and leads in 4 of 7 scored categories. Agent Platform Memory Bank leads on security & auth and transparency & trust.

Do Agent Platform Memory Bank and past.dev need an API key?

Agent Platform Memory Bank uses an OAuth sign-in. past.dev takes an API key or an OAuth sign-in.

Can an agent call Agent Platform Memory Bank and past.dev without installing anything?

Yes. Agent Platform Memory Bank has a hosted endpoint at https://aiplatform.googleapis.com/v1 and past.dev at https://api.past.dev.

Other comparisons with Agent Platform Memory Bank or past.dev

Machine-readable

For companies

Do agents find, use and choose your tools?

An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.