A1 ratingsDatabricksMosaic AI Agent Framework

Mosaic AI Agent Framework

Databricks · framework · Any industry
A
AGrade
72Score
±1295% range
≈#46Rank of 1134
0%From fleets
FR-CAT-00326Catalog no.

The six pillars

Each bar is the score; the bracket is its 95% range.

Can you see what it costs?

Cloud billing export (system.billing.usage (+ system.billing.list_prices)) · prices published
69±27
Desk research 76 · counts for 25% · A cost API you can call scores highest; an admin console or cloud billing export next; a card statement or emailed invoice is all some products leave you. Published prices add to it.

Can you see how much it's used?

system.billing.usage
86±27
Desk research 100 · counts for 15% · A usage API (tokens, seats, credits, conversations) a buyer can read.

Can you tie cost to one agent?

By project
64±27
Desk research 70 · counts for 15% · Exact (one key or project per agent) beats by-project, by-workspace or by-user.

Can an agent on it report out?

Remote MCP connector · Databricks hosts managed MCP servers (Genie, AI Search, DBSQL, UC functions); agent can also query system.billing.usage via SQL/SDK.
86±27
Desk research 100 · counts for 15% · A remote MCP connector scores highest, then a local MCP server, an SDK, an API or a workflow step; copy-and-paste lowest. Any open standard counts, not just FleetRegistry.

Can you control spend and access?

Spend cap: Hard cap · Per-agent keys: Scoped · Audit log: By API · Revoke: By API · SSO: SSO + SCIM
86±27
Desk research 100 · counts for 20% · Hard spend caps (30%), scoped per-agent keys (25%), an audit log you can read by API (20%), revoking by API (15%), SSO with SCIM (10%).

Does it work on small plans?

Small plans don't get the data
14±27
Desk research 0 · counts for 10% · Whether the plan a person or small business buys gets the cost and usage data at all.

Where the evidence comes from

Independent fleets0%Desk research71%Neutral starting point29%

Once three independent fleets report on a pillar, desk research counts for at most a third of it. FleetRegistry's own fleets never count.

Controls on record

Spend capHard cap
Per-agent keysScoped
Audit logBy API
RevokeBy API
SSOSSO + SCIM
PricesPublished

Budgets alert by default; 'Block usage' exists only for Unity Gateway (LLM/agent requests) and is enforced approximately, and Databricks says not to rely on it as an absolute cap; provisioned throughput and compute budgets are alert-only. Token permissions, auto-scoped/applied token scopes and the Token Management API (list/delete) give scoped revocable identities; audit via system.access.audit (Public Preview); SCIM requires Premium or above.

History

11-12 A 7211-13 A 7211-14 A 7211-15 A 7211-16 A 7211-17 A 7211-18 A 7211-19 A 72
Rated since 2026-10-02. Published grades are sealed daily into the transparency log (N4.4). Grade held since 2026-10-02.

Every fact, with its rung and source

Each fact stands on its own source and its own rung of the evidence ladder (Rule N2.1). Verified Observed Declared Inferred. The weakest fact in a pillar sets its weight.

FactValueRungCheckedSources
How a buyer sees cost
Cost
Cloud billing export Declared
medium confidence
2026-10-25
1 earlier
[1][2][3][4] Correct
Prices published
Cost
Yes Declared
medium confidence
2026-10-25
1 earlier
[1][2][3][4] Correct
Usage API
Usage
system.billing.usage Declared
medium confidence
2026-10-25
1 earlier
[1][2][3][4] Correct
Cost per agent
Per agent
By project Declared
medium confidence
2026-10-25
1 earlier
[1][2][3][4] Correct
How an agent reports out
Reports out
Remote MCP connector Declared
medium confidence
2026-10-25
1 earlier
[1][2][3][4] Correct
Per-agent keys
Controls
Scoped Declared
medium confidence
2026-10-25
1 earlier
[1][2][3][4] Correct
Audit log
Controls
By API Declared
medium confidence
2026-10-25
1 earlier
[1][2][3][4] Correct
Revoke access
Controls
By API Declared
medium confidence
2026-10-25
1 earlier
[1][2][3][4] Correct
Spend cap
Controls
Hard cap Declared
medium confidence
2026-10-25
1 earlier
[1][2][3][4] Correct
SSO
Controls
SSO + SCIM Declared
medium confidence
2026-10-25
1 earlier
[1][2][3][4] Correct
Small plans get the data
Small plans
No Declared
medium confidence
2026-10-25
1 earlier
[1][2][3][4] Correct

Prices · DBU-based, rate varies by SKU (not read)

Notes · Model serving endpoints billed via DBUs; system.billing.usage carries usage_metadata.endpoint_id/endpoint_name (serving), app_id (Apps) and agent_bricks_id (Knowledge Assistant/Supervisor Agent), so cost attributes per endpoint/agent. Databricks' product page is now titled 'Agent Bricks' and docs call it the agent developer platform; no page we read states that Mosaic AI Agent Framework was formally renamed.

Scored with the rubric for Open-source frameworks (Rule N7.1). Each pillar blends a neutral 50, desk facts weighted by rung, and independent fleets. A1 is 85+ with fleet proof, A 70, B 55, C 40, D 25, E below. Part N of the Rules

Run agents on Mosaic AI Agent Framework?

Give each one a permanent number, a named person who answers for it, and a record of what it can touch and what it costs. Free.