Juju: Infrastructure Automation Ledger 2026 Scorecard

Part of the Infrastructure Automation Ledger (edition 2026-10-16). How the scores work: methodology.

Snapshot

Type: product
Owner: Canonical Ltd (UK, privately held)
Licence: AGPLv3; free; no Juju-specific price found; JAAS (central control plane for many controllers) offered without a published price
Latest release: 4.0.15 (2026-09-23); 3.6.29 LTS (2026-09-30)
Support lifecycle: 3.6 LTS to 2039-04-11 on the current table (an older roadmap page said 2030-05-30); 2.9 LTS to 2035-04-28; 3.0 to 3.5 each lived under a year; 4.0 (2025-11-14) has no end of life listed
Next milestone: none published; the 4.0 notes defer a lighter event API and LXD profiles for Kubernetes workloads without dates
Targets: Machines and Kubernetes on AWS, Azure, Google Cloud, OCI, OpenStack, vSphere, LXD and MAAS bare metal (vendor list; the Equinix Metal entry is stale); KVM containers removed in 4.0; no Proxmox VE, Nutanix, Hyper-V, XCP-ng or KubeVirt; Terraform provider 2.4.1 (2026-10-08)
Tier: Host and bare-metal provisioning
Archetype: Slow follower
Evidence grade: C. Versions, the lifecycle table and the 4.0.0 release notes were read from Canonical pages and GitHub. The cloud and hypervisor list is a vendor claim from the product page. No pricing was found. Releases before 2.9 (2021) were not found, so most capability lines are held flat from 2015 on inferred values.

Scope of the scores: Juju manages charmed applications more than hosts: its unit of work is an application and its relations, and machines are provisioned to carry them. It passed inclusion test 4 only partially: bundles and diff-bundle give a desired state and a comparison against a live model, but there is no saved plan and 4.0 removed export-bundle. Scores read it as an application operator that also provisions.

Scorecard

Dimension	Score	Direction
Capability today	49.6 / 100	higher is better
Momentum since end of 2022	+0.8 points (2015 to 2022: +0.0)	higher is better
Credibility	14.5 / 25	higher is better
Commercial risk	6.5 / 25	lower is better
Technical lock-in	12.5 / 25	lower is better
Integrator fit	10.5 / 20	higher is better
AI leverage	2.0 / 20	higher is better
Ledger Index	47.8 / 100	balanced weights; see the methodology

Capability by domain, 2015 to today

Scores 0 to 5 against fixed anchors; total is weighted to 100.

Domain (weight)	2015	2019	2022	Today
Provisioning breadth (15)	3	3	3	3
Configuration management depth (12)	2.5	2.5	2.5	2.5
State, planning and safety (12)	2	2	2	2
Language and authoring (10)	3	3	3	3
Orchestration and self-service (10)	3	3	3	3
Scale and control plane (8)	2.5	2.5	2.5	3
Security and compliance (10)	2.5	2.5	2.5	2.5
Bare metal and day-2 operations (8)	2	2	2	2
Ecosystem and integrations (8)	3	3	3	3
AI assistance (7)	0	0	0	0
Total / 100	48.8	48.8	48.8	49.6

Year by year: 2015: 48.8 · 2016: 48.8 · 2017: 48.8 · 2018: 48.8 · 2019: 48.8 · 2020: 48.8 · 2021: 48.8 · 2022: 48.8 · 2023: 48.8 · 2024: 48.8 · 2025: 48.4 · 2026: 49.6

What moved the score

Year	Domain	Change	Points	Trigger
2025	Scale and control plane	2.5 → 3	+0.8	4.0: Dqlite relational controller database; HA by scaling controller units
2025	State, planning and safety	2 → 1.5	-1.2	4.0: export-bundle removed
2026	State, planning and safety	1.5 → 2	+1.2	terraform-provider-juju 2.4: plan for models and applications; diff-bundle compares a bundle with a live model

Direction, 2023 to 2026

Each release item carries one theme and one driver (V vision, C customer need, M market window, U upstream, P portfolio).

Date	Item	Theme	Driver
2023-02	3.1: first of three 3.x minors in 2023, each supported under a year	Packaging	P
2024-05	3.5: last short-lived minor before the LTS (supported 2024-05 to 2024-11)	Packaging	P
2024-12	3.6 designated LTS with security fixes to 2039-04-11	Licensing	C
2025-11	4.0: Dqlite relational controller database; HA by scaling controller units	Scale	V
2025-11	4.0: export-bundle, KVM containers, status watch and leader settings removed; podspec charms no longer run	State and planning	P
2025-11	4.0: bases replace series; manual provider renamed unmanaged	Language	U
2025-11	4.0: lighter event API to replace AllWatcher (planned, no date)	Orchestration	U
2026-01	4.0.1 to 4.0.15: patch releases roughly monthly through 2026	Packaging	C
2026-09	3.6.29: LTS line maintained in parallel	Packaging	C
2026-10	terraform-provider-juju 2.4.1	State and planning	C

Centre of gravity: Packaging (4), State and planning (2), Licensing (1)
Blind spots (no items): Provisioning breadth, Configuration management, Security, Bare metal, Ecosystem, AI
Driver mix: Vision 10%, Customer need 40%, Market window 0%, Upstream 20%, Portfolio 30%
Idea-to-need index: 0.20 (0 = demand-pulled, 1 = thesis-pushed)

Credibility: 14.5 / 25

Criterion	Score	Basis
Cadence	3.5	Frequent patch releases on 3.6 and 4.0; minors every 3 to 12 months; dates not published in advance
Roadmap transparency	2.5	Release notes public; roadmap page not dated; the 4.0 notes list deferred items without dates
Say-do	3	Few announcements to check; 3.6 LTS and 4.0 shipped; the lighter event API is pending
Velocity	3	4.0 rebuilt the controller store and removed several commands; patch releases are maintenance
Lifecycle stability	2.5	3.6 LTS to 2039 is long, but 3.0 to 3.5 each lived under a year, 4.0 needs a model migration from 3.6 and has no published end of life

Commercial risk: 6.5 / 25 (lower is better)

Criterion	Score	Basis
Price and licence volatility	1	AGPLv3, free; no Juju price found and no change found
Purchase constraints	1	No Juju SKU; support comes through Canonical contracts
Channel and access	1	Open source, snaps and apt; Charmhub is public
Owner stability	1.5	Single privately held vendor controls the project
Cost of staying supported	2	Moving from 3.6 to 4.0 requires model migration; 4.0 has no published end of life

Technical lock-in: 12.5 / 25 (lower is better)

Criterion	Score	Basis
Formats	2.5	Charm and bundle formats are specific to Juju
Export path	3	export-bundle removed in 4.0; no generic export
Stack coupling	2.5	Needs a Juju controller and charms; application knowledge lives in charms
Hardware / cloud coupling	1	Many clouds; Ubuntu focus
Skills and tooling coupling	3.5	Charm authoring (ops framework, Python) is a scarce skill

Integrator fit: 10.5 / 20

Criterion	Score	Basis
Partner programme openness	2	Open source; Canonical partner terms not researched
Multi-tenancy and self-service	3	RBAC and multi-model controllers; JAAS for many controllers (vendor claim)
API and infrastructure-as-code	3.5	Full API and CLI; Terraform provider 2.4.1
Skills and certification	2	Charm skills are scarce; Canonical training not verified

AI leverage: 2.0 / 20

How the product uses AI, scored per year like capability. Only GA counts.

Criterion	2019	2022	Today	Latest step
AI code and plan generation	0	0	0	2015: None
AI platform stack	0	0	0	2015: None
AI-assisted operations	0	0	0	2015: None
Agent openness	2	2	2	2015: API and CLI; the mcp-juju snap on the edge channel is unofficial and its publisher is not confirmed

Year by year: 2015: 2 · 2016: 2 · 2017: 2 · 2018: 2 · 2019: 2 · 2020: 2 · 2021: 2 · 2022: 2 · 2023: 2 · 2024: 2 · 2025: 2 · 2026: 2

AI say-do, 2023 to 2026: 0 AI items announced; 0 GA, 0 still preview, 0 pending, 0 dropped. No official AI items; an unofficial mcp-juju snap sits on the edge channel at version 0.0.12.

Juju has no AI generation and no AI in its operations. Agents can drive the API and CLI; the only MCP server is an unofficial snap on the edge channel.

Reading

Juju's capability total looks respectable because it covers many clouds and brings a real language, registry and orchestration model, but those strengths belong to application operations rather than infrastructure provisioning, which is why it carries a scope note. The 4.0 line is a controller rewrite that removed export-bundle, KVM containers and other commands, and that removal costs a half step in State, planning and safety until the Terraform provider restored a plan path. Commercial risk is low because nothing is sold separately; lock-in is the highest in the group because knowledge lives in charms. Check which line you are buying into: 3.6 is supported to 2039, 4.0 has no published end of life.

Open questions and items to re-score

The vSphere and other provider claims come from the product page; provider documentation was not read.
Whether 4.0 has any plan or dry-run beyond diff-bundle; deploy --dry-run is documented only for the 2.0 era.
Juju pricing and support terms, and JAAS pricing, were not found.
The 3.6 LTS end date conflicts between the current table (2039-04-11) and an older roadmap page (2030-05-30).
End of life for 4.0 is not listed.
The 1.x and 2.x release history (2015 to 2020) was not found; early scores are inferred.

	’25	’26
Provisioning breadth
Configuration management depth
State, planning and safety		1
Language and authoring
Orchestration and self-service
Scale and control plane	1
Security and compliance
Bare metal and day-2 operations
Ecosystem and integrations
AI assistance