Compare / Meta Muse vs Town

Meta Muse vs Town

Meta Muse is a persistent personal agent from Meta (United States); Town is a work-based personal agents and team automation from Town.com (United States). On Assistant Benchmark (1–10, updated 8 Oct 2026), Meta Muse scores 9.3 over 8 dimensions and Town scores 8.3 over 7; both were scored on 5 of the same dimensions, shown below. PAEval holds 30 capability records for Meta Muse and 15 for Town (as of 9 Oct 2026).

Scores stay on each evaluator's own scale and are never combined. PAEval does not pick a winner.

Meta MuseTown
VendorMetaTown.com
HQ regionUnited StatesUnited States
TypePersistent personal agentWork-based personal agents and team automation
Pricing (as documented)Free has a limit; Power is US$20/month, 500M Muse tokens/week; Maximum is US$100/month, 3B/week; automatic monthly renewal.Not compiled
Assistant Benchmark1–10 mean of scored dimensions9.3coverage 8/158.3coverage 7/15
micro1 PersonalAgentBench% of 10 workflows with trusted completion20%95% CI 6–51%Not scored
SuperCLUE-XClaw0–100 weighted total of 5 dimensionsNot scoredNot scored
Autonomy & background workDocumentedDocumented
Execution environmentsDocumentedPartly disclosed / partly tested
Calls, messages & identityLimited or conditionalDocumented
Connectors & toolsDocumentedDocumented
Memory & personalizationDocumentedDocumented
Channels & interfacesDocumentedDocumented
Research & deliverablesDocumentedDocumented
Transactions & bookingsLimited or conditionalNo record
Collaboration & parallelismDocumentedDocumented
Reliability & recoveryLimited or conditionalDocumented
Permissions, security & dataLimited or conditionalDocumented
Capability records30 (0 independent)15 (1 independent)
Capability cells show the strongest documented or observed status in each domain; availability is not task success.Add more products →

Assistant Benchmark, dimension by dimension

Meta MuseTown
Carrying out an online task9testNot tested
Travel booking7observed in useNot tested
Recommendation quality9test8test
Purchasing a product10observed in useNot tested
Responding to emails10observed in use8test
Proactive behaviorNot testedNot tested
Running a routine10observed in use7test
Third-party integrations10observed in use10observed in use
Permissions & privacy9observed in use8observed in use
MemoryNot tested8unlabelled
Phone callsNot testedNot tested
Multiplayer / groupsNot testedNot tested
Chained tasksNot testedNot tested
Proactive restraintNot testedNot tested
Content creation / gamesNot tested9test
Source scores on the 1–10 anchor scale. "Not tested" is missing evidence, not zero.Source ↗