Knowledge & context
Which AI agent platforms remember context across runs and conversations?
Memory is the scarcest of the core capabilities. Of 946 vendors, only 102 document full persistence of context across a run, a conversation, a workflow, a user, a team or a long term memory layer. 418 document partial coverage and 426 document none.
Every vendor in the index is assessed against the same 14 point taxonomy from public documentation, and no vendor pays for placement. Counts on this page were measured across all 946 public vendors on September 30, 2026.
How the 946 vendors split
No public evidence means the reviewed sources did not document the capability. On this index that is a statement about the evidence, not proof that the capability is absent. See methodology.
What counts as full coverage
Full coverage means persistence the buyer can reason about: a documented memory layer with a stated scope and lifetime, whether that is per user, per team or per workflow. Partial almost always means conversation state inside a single session, which every chat interface has and which disappears the moment the session ends. Learning across sessions is not a memory layer, and neither is the application's own database: if deleting the memory would delete the business record, it is the data model, not memory.
How to read these numbers
Multi agent platforms lead at 49 percent, because coordinating several agents forces shared state to become an explicit product feature, and coding follows at 44 percent. Security sits at 11 percent and healthcare at 12 percent, which is worth reading twice: those are the lanes where remembering across encounters carries the most operational value and the most retention risk, and the documentation is among the thinnest exactly there. Buyers should ask two questions the marketing rarely answers. Where does the memory live, and how do you delete it for one customer without wiping the rest.
Leading platform for memory & state persistence in each use case
Picked mechanically: the highest total coverage vendor in each lane that documents full evidence on this axis, one vendor per row. Scores are out of 14.
-
1. Appian, for enterprise operations agents
14 / 14Full enterprise operations agents ranking · Compare the whole lane on all 14 axes
-
2. FLOWX.AI, for multi-agent platforms
14 / 14Full multi-agent platforms ranking · Compare the whole lane on all 14 axes
-
3. Gumloop, for GTM and revenue agents
14 / 14Full GTM and revenue agents ranking · Compare the whole lane on all 14 axes
-
4. Mastra, for agent infrastructure platforms
14 / 14Full agent infrastructure platforms ranking · Compare the whole lane on all 14 axes
-
5. ServiceNow, for customer support agents
14 / 14Full customer support agents ranking · Compare the whole lane on all 14 axes
-
6. UiPath, for agent builders
14 / 14Full agent builders ranking · Compare the whole lane on all 14 axes
-
7. GitHub Copilot, for coding agents
13.5 / 14Full coding agents ranking · Compare the whole lane on all 14 axes
-
8. Dataiku, for data analyst agents
13 / 14Full data analyst agents ranking · Compare the whole lane on all 14 axes
-
9. HappyRobot, for voice agents
13 / 14Full voice agents ranking · Compare the whole lane on all 14 axes
-
10. Edge Delta, for SRE and DevOps agents
12 / 14Full SRE and DevOps agents ranking · Compare the whole lane on all 14 axes
-
11. Viktor, for browser and computer-use agents
11 / 14Full browser and computer-use agents ranking · Compare the whole lane on all 14 axes
-
12. Command Zero, for security and SOC agents
10.5 / 14Full security and SOC agents ranking · Compare the whole lane on all 14 axes
Documented coverage by use case
Share of each lane documenting full coverage on this axis. Vendors that sit in two lanes count in both, the same rule the rankings and matrices use.
| Use case | Full coverage | Share |
|---|---|---|
| multi-agent platforms | 21 of 52 | 40% |
| coding agents | 16 of 64 | 25% |
| agent builders | 23 of 115 | 20% |
| SRE and DevOps agents | 6 of 37 | 16% |
| agent infrastructure platforms | 28 of 186 | 15% |
| data analyst agents | 8 of 59 | 14% |
| customer support agents | 13 of 113 | 12% |
| voice agents | 7 of 83 | 8% |
| GTM and revenue agents | 9 of 138 | 7% |
| enterprise operations agents | 22 of 297 | 7% |
| browser and computer-use agents | 3 of 42 | 7% |
| security and SOC agents | 2 of 85 | 2% |
| healthcare agents | 0 of 69 | 0% |
Recent verified changes from the vendors named above
Capability coverage is not a static picture. These are the most recent sourced change log entries for the platforms listed above, newest first, one per vendor. Scores on this page update as entries like these are verified.
-
GitHub Copilot memory / state
Medium impactGitHub released a batch of updates for Copilot across desktop, CLI, and VS Code. Key additions include an experimental worktree command for isolated worktrees, a rewind command to restore conversations without Git, session management in the CLI, and multilingual on-device dictation in VS Code.
August 7, 2026 · Verified · All GitHub Copilot changes
-
Mastra memory / state
Medium impactMastra released Memory Extractors, a feature that enables agents to automatically pull and persist structured data from natural language conversations.
July 21, 2026 · Partially Verified · All Mastra changes
-
UiPath human approval / guardrails
High impactAdministrators can now bring their own safety vendor into UiPath's AI Trust Layer, in preview, so agent guardrails run under the customer's own vendor agreement and region. Azure AI Language, Azure Content Safety and Noma are supported, checks can be enforced across the organization, and each evaluation is recorded in the agent trace.
October 1, 2026 · Partially Verified · All UiPath changes
-
Gumloop agent capability
Medium impactUsers can now text Gumball, Gumloop's personal agent, over iMessage, RCS or SMS to start a task, get replies in the same thread and approve steps by text. It is a public beta that admins can switch off by role.
September 29, 2026 · Partially Verified · All Gumloop changes
-
FLOWX.AI agent capability
High impactFlowX.AI 5.13.0 adds an AI agent to its Designer that surveys an existing app, drafts a plan from a plain language request, then builds or edits processes, screens, workflows and data types. It runs only after the builder confirms the plan and reviews each step, works on its own branch, and refuses destructive changes such as deleting a process with live instances.
September 28, 2026 · Verified · All FLOWX.AI changes
Full change log · updated weekly across the whole index
Questions buyers ask
Why is session context only partial credit?
Because it is not persistence. Holding a conversation together inside one session is table stakes. Memory that survives the session is what changes how an agent behaves on the second interaction.
Is memory a retention risk?
It can be. Persisted context is often personal or commercially sensitive, so pair this axis with deployment and data residency and with security and identity before enabling long term memory in a regulated setting.
Which lane documents memory best?
Multi agent platforms, at roughly one in two members. GTM, voice and enterprise operations are the thinnest, each under one in ten, with security and healthcare just above that.
The other 13 axes
No single axis decides a shortlist. Buyers who care about this one usually check observability & auditability and deployment & data residency next, or open the full taxonomy to see how the 14 axes fit together.