SeaweedFS: Object Storage Ledger 2026 Scorecard

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

Snapshot

Type: product (open source project with a paid Enterprise Edition)
Owner: the seaweedfs GitHub organisation, an independent Apache-licensed project funded through Patreon and sponsors (README); the Enterprise Edition is sold by SeaweedFS Enterprise, whose legal entity and country were not found
Licence: Apache 2.0. Enterprise Edition: a licence key added to an existing open source cluster, free for development and test below 25 TB, paid at $2 per TB per month or $20 per TB per year (seaweedfs.com); how the rate applies in production above 25 TB is not stated
Latest release: 4.48 (2026-09-28)
Support lifecycle: none published; rolling releases about once a week (4.41 on 2026-08-06 to 4.48 on 2026-09-28). Enterprise support by email, Slack and Telegram with a response within 24 hours on business days
Next milestone: none published
Object lock: Retention and legal hold listed in the README; Object Lock fixes in 4.45, 4.46 and 4.48 (releases). Governance and compliance modes were not confirmed in sources read, and no backup-vendor validation was found
Tier: Open-source software
Archetype: Dependable builder, open core
Evidence grade: B. Releases 4.41 to 4.48, the README and the seaweedfs.com Enterprise page were read in October 2026. The releases page shows dates without a year; 2026 is inferred from asset timestamps and the repository push date. Release history from 2015 to 2025 was not read, so the steps before 2026 and their years are inferred (grade C).

Scorecard

Dimension	Score	Direction
Capability today	69.3 / 100	higher is better
Momentum since end of 2022	+15.3 points (2015 to 2022: +29.3)	higher is better
Credibility	16.0 / 25	higher is better
Commercial risk	8.0 / 25	lower is better
Technical lock-in	7.0 / 25	lower is better
Integrator fit	8.5 / 20	higher is better
AI leverage	5.0 / 20	higher is better
Ledger Index	61.6 / 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
S3 API fidelity (15)	0	2	3	4
Immutability and compliance (12)	0	0	0	2.5
Durability and resilience (12)	1.5	2.5	2.5	3
Performance and scale (12)	3	3.5	4	4
Deployment freedom (10)	3	3.5	4	4
Security and multi-tenancy (10)	0.5	1.5	2.5	3
Operations (10)	1	2	2	3
Protocols beyond S3 (7)	1	3	3	4
Lifecycle and tiering (7)	0	2	3	3.5
Ecosystem and openness (5)	3.5	4	4	4
Total / 100	24.7	45.4	54.0	69.3

Year by year: 2015: 24.7 · 2016: 24.7 · 2017: 24.7 · 2018: 24.7 · 2019: 45.4 · 2020: 45.4 · 2021: 45.4 · 2022: 54.0 · 2023: 54.0 · 2024: 54.0 · 2025: 54.0 · 2026: 69.3

What moved the score

Year	Domain	Change	Points	Trigger
2019	Deployment freedom	3 → 3.5	+1.0	Docker images and Helm chart; year approximate (grade C)
2019	Durability and resilience	1.5 → 2.5	+2.4	Erasure coding for warm volumes beside replication for hot data; year approximate (grade C)
2019	Ecosystem and openness	3.5 → 4	+0.5	Apache 2.0 without redistribution limits, outside contributors, Docker, Helm and CSI; few named backup-vendor integrations; year approximate (grade C)
2019	Lifecycle and tiering	0 → 2	+2.8	Warm volumes tiered to cloud storage; year approximate (grade C)
2019	Operations	1 → 2	+2.0	Prometheus metrics and weed shell volume operations; year approximate (grade C)
2019	Performance and scale	3 → 3.5	+1.2	Filer with pluggable metadata stores (PostgreSQL, MySQL, Redis, etcd and others), so metadata scales apart from capacity; year approximate (grade C)
2019	Protocols beyond S3	1 → 3	+2.8	Filer with FUSE mount, WebDAV and a Hadoop-compatible client; year approximate (grade C)
2019	S3 API fidelity	0 → 2	+6.0	S3 gateway with buckets, multipart and basic object operations; year approximate (grade C)
2019	Security and multi-tenancy	0.5 → 1.5	+2.0	S3 access keys with per-bucket actions; year approximate (grade C)
2022	Deployment freedom	3.5 → 4	+1.0	SeaweedFS Operator and CSI driver; single-binary weed mini for small sites; year approximate (grade C). Arm builds not verified
2022	Lifecycle and tiering	2 → 3	+1.4	S3 lifecycle expiration rules; year approximate (grade C)
2022	Performance and scale	3.5 → 4	+1.2	Billions of files claimed by the project; no independent benchmark and no RDMA; year approximate (grade C)
2022	S3 API fidelity	2 → 3	+3.0	Versioning, lifecycle expiration, tagging, bucket policies and SSE; year approximate (grade C)
2022	Security and multi-tenancy	1.5 → 2.5	+2.0	IAM API, bucket policies and server-side encryption; year approximate (grade C)
2026	Durability and resilience	2.5 → 3	+1.2	Enterprise Edition: automatic EC repair, EC bitrot scrub, self-healing storage, custom EC ratio (launch date not found); 4.42 EC rollback and cleanup
2026	Immutability and compliance	0 → 2.5	+6.0	Retention and legal hold in the README; Object Lock fixes in 4.45, 4.46 and 4.48; first release with Object Lock not established, so the step sits in the first year verified; no backup-vendor validation
2026	Lifecycle and tiering	3 → 3.5	+0.7	4.41 bounded lifecycle replay, 4.44 lifecycle editing in the admin UI, 4.47 bucket quotas; no archive class with restore found
2026	Operations	2 → 3	+2.0	Admin UI with bucket lifecycle editing (4.44) and a visual IAM policy editor (4.45); no published upgrade path or support window
2026	Protocols beyond S3	3 → 4	+1.4	Built-in Iceberg REST catalog with S3 Tables and Lance tables (2026); 4.42 Iceberg compare-and-swap commits and vended credentials; SFTP; no NFS or SMB
2026	S3 API fidelity	3 → 4	+3.0	README lists 73 object and 36 bucket operations, 5 STS operations, SSE-S3, SSE-KMS and SSE-C, S3 Tables; 4.42 RenameObject, 4.48 versioned POST policy uploads
2026	Security and multi-tenancy	2.5 → 3	+1.0	STS operations in the open source endpoint; Enterprise admin UI with OIDC, multi-tenancy and S3 QoS; 4.47 explicit Deny precedence in bucket policies, attached IAM policies protected from deletion

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
2026	Enterprise Edition by licence key at $2 per TB per month, free for development and test below 25 TB (launch date not found)	Packaging	P
2026-04	Built-in Iceberg REST catalog and S3 Tables (README; first release not dated)	Protocols	M
2026-04	README states MinIO ceased development and lists alternatives	Ecosystem	M
2026-08	4.41: S3 manifest chunks for large writes, suspended-versioning delete markers, bounded lifecycle replay	S3 API	C
2026-08	4.42: master heartbeat overhaul, EC rollback and cleanup, S3 RenameObject, Iceberg compare-and-swap commits	Durability	C
2026-08	4.43: Lance catalog, Rust plugin worker, Go 1.26 support	AI	V
2026-08	4.44: bucket lifecycle editing in the admin UI	Operations	C
2026-08	4.45: visual IAM policy editor; Object Lock covers directory-marker keys	Immutability	C
2026-09	4.46: Object Lock on directory-marker keys; Docker images signed with cosign	Security	C
2026-09	4.47: bucket quota extension, explicit Deny precedence, attached IAM policies protected from deletion	Security	C
2026-09	4.48: Object Lock configuration permission and version deletes, versioned POST policy uploads, EC fixes	Immutability	C

Centre of gravity: Immutability (2), Security (2), Packaging (1)
Blind spots (no items): Performance, Deployment, Lifecycle, Licensing
Driver mix: Vision 9%, Customer need 64%, Market window 18%, Upstream 0%, Portfolio 9%
Idea-to-need index: 0.10 (0 = demand-pulled, 1 = thesis-pushed)

Credibility: 16.0 / 25

Criterion	Score	Basis
Cadence	4	Tagged releases about once a week in August and September 2026; no published rhythm
Roadmap transparency	3	Public release notes and an open issue tracker; no roadmap page
Say-do	3	No dated promises to test; features appear in release notes when they ship
Velocity	4	Several items per release in 4.41 to 4.48, many of them fixes
Lifecycle stability	2	Rolling versions with no support window; each fix means moving to the newest build

Commercial risk: 8.0 / 25 (lower is better)

Criterion	Score	Basis
Price and licence volatility	1	Apache 2.0 with no relicensing found; Enterprise price of $2 per TB per month read in 2026, launch date and earlier prices not found
Purchase constraints	1.5	Open source build needs no purchase; Enterprise free below 25 TB for development and test, terms for production above that not stated
Channel and access	1	Open download through GitHub, Docker and Helm; Enterprise by licence key and EULA
Owner stability	2.5	Funded through Patreon and four gold sponsors (README); the legal entity and country behind SeaweedFS Enterprise were not found
Cost of staying supported	2	No support window; EC repair, bitrot scrub and self-healing are Enterprise-only; weekly releases mean frequent upgrades

Technical lock-in: 7.0 / 25 (lower is better)

Criterion	Score	Basis
Formats	2	Append-only volume files that need SeaweedFS to read; filer metadata in a standard database
Export path	1	S3, FUSE mount, WebDAV, SFTP and HDFS; data copies out with standard tools
Stack coupling	2	Master, volume servers, filer and S3 gateway, plus a metadata store to run
Hardware / cloud coupling	0	Any hardware
Skills and tooling coupling	2	SeaweedFS-specific operations and weed shell; smaller skills pool than Ceph

Integrator fit: 8.5 / 20

Criterion	Score	Basis
Partner programme openness	1.5	No partner programme found; Apache 2.0 lets integrators package and support it freely
Multi-tenancy and self-service	2.5	IAM and STS in the open source build; multi-tenancy and S3 QoS are Enterprise features (vendor page)
API and infrastructure-as-code	3	Helm, Operator, CSI driver and admin API; Terraform provider not found
Skills and certification	1.5	No training or certification found

AI leverage: 5.0 / 20

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

Criterion	2019	2022	Today	Latest step
AI workload support	1	1	1	2015: S3 and FUSE access usable as a training data source; no GPUDirect or RDMA found
AI platform stack	0	0	2	2026: Built-in Iceberg REST catalog with compaction, snapshot expiry and orphan file removal, S3 Tables and Lance tables (README; maturity not stated)
AI-assisted operations	0	0	0	2015: None
Agent openness	2	2	2	2015: HTTP, filer and S3 APIs that agents can drive; no MCP server found

Year by year: 2015: 3 · 2016: 3 · 2017: 3 · 2018: 3 · 2019: 3 · 2020: 3 · 2021: 3 · 2022: 3 · 2023: 3 · 2024: 3 · 2025: 3 · 2026: 5

AI say-do, 2023 to 2026: 2 AI items announced; 2 GA, 0 still preview, 0 pending, 0 dropped. The Iceberg catalog with S3 Tables and the Lance catalog appear in the README and release notes as shipped features; no dated AI announcement exists.

SeaweedFS adds AI-adjacent data formats to the store itself: an Iceberg catalog and Lance tables, both shipped without a launch announcement. It has no GPU data path, no AI in its own operations and no agent interface beyond its APIs.

Reading

SeaweedFS scores well on protocols and deployment because one Apache 2.0 code base serves S3, FUSE, WebDAV, SFTP, HDFS clients and an Iceberg catalog, from a single binary up to a Kubernetes Operator. Immutability is held to 2.5 because Object Lock has no backup-vendor validation on record and was still being fixed in three of the eight releases since August 2026; durability is held to 3 because automatic EC repair, bitrot scrub and self-healing sit in the paid Enterprise Edition. Credibility rests on a weekly release rhythm with no support windows, and the scores before 2026 are inferred because the older release history was not read. A buyer planning it as a backup target should test object lock with their own backup software and get SeaweedFS Enterprise's production price and support terms in writing.

Open questions and items to re-score

First release with an S3 gateway and first release with Object Lock; whether both governance and compliance modes are supported.
Backup-vendor validation (Veeam, Proxmox Backup Server, others): none found.
Enterprise Edition launch date, legal entity, country, and the production price above 25 TB.
Release history from 2015 to 2025 was not read; the years of the earlier steps are approximate.
Arm builds and the failure behaviour of the filer metadata store at scale were not verified.
The releases page shows dates without a year; 2026 is inferred.

	’19	’22	’26
S3 API fidelity	1	1	2
Immutability and compliance			1
Durability and resilience
Performance and scale
Deployment freedom
Security and multi-tenancy		2
Operations			3
Protocols beyond S3	2
Lifecycle and tiering	3	3
Ecosystem and openness