Home / Microsoft
Azure DevOps Repos (Git as bulk blob store)
azure.microsoft.com/en-us/pricing/details/devops/azure-devopPaaSstorage/azure-devops-reposJSONcomparereport an errorprovider JSON
- Free allowance
- Unlimited private Git repositories free per organization; each repository may be up to 250 GB; storage is NOT a billed meter - the Azure DevOps free tier bills only users, parallel jobs and Artifacts storage.
- Verdict
- ★ qualifiesusable as stated; the read path was probed
- Cost
- free tier
- Identity
- anonymousno card, no phone
- Binding limit
- noneno named constraint binds
- Evidence
- reported6 sources; a third party said so
- Freshness
- stale on 2026-11-05checked 2026-10-21 (inherited); stale on 2026-11-05
| Free tier | Unlimited private Git repositories free per organization; each repository may be up to 250 GB; storage is NOT a billed meter - the Azure DevOps free tier bills only users, parallel jobs and Artifacts storage. Git LFS is free and does not count toward the 5 GB push limit. unit GB; metered on stored-bytes |
|---|
| Price | none stated |
|---|
| Identity | anonymous not published |
|---|
| Egress | Unmetered. Clone/fetch over HTTPS from dev.azure.com has no egress meter - the billing FAQ enumerates every chargeable meter (extra Basic users, Basic+Test Plans, parallel jobs, Artifacts >2 GB, GHAS) and data transfer is not among them. |
|---|
| Per-object cap | not stated |
|---|
| Retention | not stated |
|---|
| Interface | rest, git, rclone part of the text of Free tier, above |
|---|
| Tooling | Plain git / git-lfs; Azure DevOps REST API for repository + item download (GET /_apis/git/repositories/{repo}/items?path=); rclone has no ADO Repos backend, so use git or the REST item API |
|---|
| Composes with | Unlimited repos x 250 GB each within one org; each org holds up to 1,000 projects; orgs are free and you may hold many. 250 GB x N repos x M orgs. The 2026 sprint-273 change removed the Git object-count limit, which was the per-repo ceiling that used to bite before 250 GB. Files are content-addressed and deduplicated across history, so a repo of distinct chunks stores each chunk once. |
|---|
| Reliability | Reliable but slow as bulk storage: no random-access read, no partial download, whole-repo clone semantics; the 'temporarily unavailable while repository is updated' pack-optimization window can block writes for large repos. |
|---|
Evidence 6
Unknown values 6
$.binding_note boilerplate$.identity_cost.detail boilerplate$.per_object_cap.notes the field may be stated in a source that was not read$.per_object_cap.stated the field may be stated in a source that was not read$.retention.notes the field may be stated in a source that was not read$.retention.stated the field may be stated in a source that was not read