airbytehq

Activity

Stars and mentions · Last 30 days, Oct 7 – Nov 5

GitHub stars116
Mentions2
036912Oct 7Oct 21Nov 5
UTC days · repositories airbytehq owns3 repositories

Top repositories

airbytehq’s repositories ranked by pull request authors, as on trending.

Loading

Contributes to

Repositories outside airbytehq’s account where airbytehq opened or reviewed pull requests in the past 365 days.

Who noticed airbytehq

Everyone who starred or forked airbytehq’s repositories since the archive began in 2011.

Notable accounts

Ordered by the stars their own repositories have received.

Biggest fans

Accounts that starred the most of these repositories.

Breakout weeks

Each repository’s best week of new stars, with the release published around it.

Mentions

3posts
Nov 61 post
  1. Airbyte Pricing vs Estuary Pricing for High-Volume CDC !cover Airbyte sells pay-per-use plans on Standard and Plus, and capacity-based plans on Pro and Enterprise Flex. Estuary charges per gigabyte and per connector instance, and counts each gigabyte twice, once on capture and once on delivery. Airbyte vs Estuary: High-Volume Pricing at a Glance At high volume, the two meters pull apart, and the entry price tells you little. Two inputs decide which bill is larger: how many destinations one source feeds, and how fresh the slowest consumer needs its data. Airbyte's credit table meters 1 GB of database data at 4 credits. Estuary's Cloud plan counts that gigabyte twice at $0.50 each time, as of October 2026. Airbyte vs Estuary Pricing Models Explained Airbyte positions Standard and Plus for smaller teams with predictable data volumes. Airbyte's pricing page pitches its enterprise plans at large organizations with higher data volumes, priced on compute capacity. Those plans carry no charges based on data volume. Once a team consumes 40 or more credits a month, Airbyte says Plus is typically more affordable than Standard. Airbyte's billing documentation says the volume it measures is likely a larger representation of your data than a direct query would show. A Full Refresh sync of an API or custom source charges for every row synced. Airbyte Cloud starts with a trial of 14 days or 400 trial credits, whichever comes first. Estuary's pricing page lists the Cloud plan at $0.50 per GB of data plus $100 per connector instance a month, as of October 2026. Estuary counts data each time it is captured and again each time it is delivered, including data produced by transformations. A derivation is not a connector instance, but the data it produces still counts as volume. An Estuary connector instance is one configured source or one configured destination, prorated by the hour. A single capture can pull from many tables and still count as one instance. A Postgres to Snowflake pipeline with 500 tables therefore uses 2 connector instances. Every GB captured and every GB delivered to a destination carries the same charge. Each active connector instance adds a flat fee. The same page lists a free tier of 10 GB a month and 2 connector instances with no credit card required. A third active source or destination exceeds the free tier. Instances three to six cost $100 a month as of October 2026. Airbyte vs Estuary Pricing Comparison Every rate in the table is as of October 2026. | Vendor | Metered unit | Published rate (as of October 2026) | Connector or connection fee | Entry cost | Latency floor | High-volume terms | |---|---|---|---|---|---|---| | Airbyte | Pay-per-use on Standard and Plus; capacity on Pro and Enterprise Flex | 4 credits per GB of database data | None | 14-day or 400-credit trial | 5-minute scheduled syncs on Pro and Enterprise Flex | Capacity enforced per region; premium rate for on-demand overflow | | Estuary | GB captured and GB delivered | $0.50 per GB captured and per GB delivered | $100 per connector instance a month, $50 after the sixth | 10 GB and 2 connector instances a month, free | Sub-second to slower schedules, set per destination | Committed-use discounts on annual plans; custom Enterprise terms | Read the connector fee and the high-volume terms first, because that is where the bills separate. Every cell in the table comes from a vendor pricing or documentation page. Airbyte vs Estuary: Sync Frequency and Latency Airbyte's CDC documentation says Airbyte does not treat CDC sources as infinite streaming sources. Airbyte’s shortest scheduled sync interval is 1 hour on Standard, 15 minutes on Plus, and 5 minutes on Pro and Enterprise Flex. Airbyte uses logical replication of the Postgres write-ahead log to incrementally capture deletes. Airbyte's Postgres connector documentation recommends that method for databases larger than 500 GB. Incremental CDC also needs primary keys on most sources, so a table without one is not a CDC candidate. Estuary charges no extra for millisecond latency or batch delivery. Delivery latency is set per destination, from sub-second to slower schedules. A single source can feed a dashboard at sub-second latency and an archive on a slower schedule. Airbyte's fastest plan still runs on a 5-minute schedule, while Estuary sets delivery latency per destination. How Airbyte and Estuary Costs Scale with Multiple Destinations In Airbyte, a connection pairs a source with a destination and syncs data from the source to the destination. Estuary supports one-to-many replication, so a single source can feed several destinations at once. Each active source or destination counts as one connector instance, prorated by the hour. Suppose a Postgres source produces 1 TB of changes a month, and both Snowflake and a search index need the full dataset. Airbyte uses two connections, each of which reads the same changes from Postgres. That doubles the source reads and metered volume compared with syncing to one destination: 2,000 GB in total, or 8,000 credits at 4 credits per GB. On Airbyte Standard, that comes to $39,995 a month: $20 for the plan, including 5 credits, plus $5 for each of the remaining 7,995 credits. This estimate is specific to Standard; Plus packages and capacity-based plans are priced differently. Airbyte Docs Estuary captures the changes from Postgres once and delivers them to both destinations. Assuming 1,000 GB captured and 1,000 GB processed for each destination, Estuary bills 3,000 GB. At Estuary’s Cloud rates, that comes to $1,800 a month: $1,500 for data and $300 for one source and two destination connectors running for the full month. A fan-out to three destinations is three connections on Airbyte and four connector instances on Estuary. What high-volume CDC pricing comes to at scale Suppose the source changes 1 TB a month into one destination. Estuary's meter counts 2 TB, once on capture and once on delivery, before the instance fee. Airbyte's bill for the same change depends on which plan runs the sync and which region runs it. Airbyte's data worker documentation says each region gets its own allocation, so one region can be full while another has headroom. When committed capacity is exhausted, syncs queue by default. Connections with on-demand capacity enabled can run immediately at a premium rate. On the credit plans, running out does not stop a sync: Airbyte keeps it running and bills in arrears. Once credit usage passes $10,000 in a billing period, Airbyte charges the saved payment method automatically, as of October 2026. Enterprise Flex adds in-boundary and hybrid deployment, including a model with an Airbyte-managed control plane. Estuary discounts committed usage on annual plans. Its Enterprise terms are custom and keyed to data volume, connector instances, deployment model, contract length, and support needs. Private and BYOC deployments require an annual or multi-year Enterprise contract. Airbyte removes the volume charge inside committed capacity and prices the overflow at a premium. On Standard and Plus, the same terabyte meters at 4 credits. Enterprise Flex meter data workers instead. Estuary's Cloud rate is $0.50 per GB as of October 2026, and its annual plans carry volume discounts on committed usage. Airbyte vs Estuary pricing: which one to pick Pick Estuary if one source has to feed several destinations, or if any consumer needs changes within seconds. Pick Airbyte if every consumer tolerates a scheduled sync. Standard and Plus suit smaller teams with predictable volumes. Pro and Enterprise Flex fit large organizations that would rather buy compute capacity than gigabytes. Either way, price the pipeline on the metered units and the destination count, not the entry price. Count the destinations, decide how fresh the slowest consumer needs its data, and read the rate for the plan you would run. Compare a credit rate against a per-GB rate counted twice, and an instance fee against a connection model. FAQ Airbyte's rate card is in credits and Estuary's in dollars. Is Airbyte free? Airbyte Open Source, the self-hosted edition, is free to use, but you cover the infrastructure costs and the engineering time needed to run and maintain it. Airbyte Cloud starts with a trial that lasts 14 days or 400 trial credits, whichever comes first. Standard and Plus are pay-per-use plans for smaller teams with predictable data volumes. Airbyte says Plus is typically more affordable than Standard at 40 or more credits a month. Pro and Enterprise Flex are capacity-based plans that use data workers. How much does a gigabyte of database data cost on Airbyte? Airbyte's credit table meters 1 GB of database data at 4 credits, as of October 2026. Its docs say that form is likely larger than what a direct query returns. The page states the rate in credits instead of dollars, so you need your plan's credit price to finish the math. Does Airbyte charge for full refreshes? Yes, for Full Refresh syncs of API and custom sources. Every row synced is charged, so re-reading a whole table bills the full row count again. Database sources are metered by volume instead. Each GB of database data costs 4 credits, as of October 2026. Is Estuary cheaper than Airbyte for high-volume CDC? It depends on the meter and the destination count, not on a single headline rate. Estuary's own example bills 100 GB a month from Postgres to Snowflake as 200 GB of data and 2 connector instances. That comes to $300 a month as of October 2026. On Airbyte, 1 GB of database data is 4 credits. Airbyte charges per credit on some plans and by capacity on Pro and Enterprise Flex. What counts as an Estuary connector instance? Every active source or destination counts as one connector instance, prorated by the hour. The table count does not change that. A single capture can pull from many tables and still count as one instance. A Postgres to Snowflake pipeline with 500 tables uses 2 connector instances. The data produced by transformations still counts toward your volume. Can you run either tool in your own cloud account? Airbyte's Enterprise Flex adds in-boundary and hybrid deployment, including a model with an Airbyte-managed control plane. Estuary's private and BYOC deployments require an annual or multi-year Enterprise contract. Both enterprise options require a contract.

  2. Airbyte connector secrets need a deployment-specific control model Airbyte connector secrets need a deployment-specific control model Data integration platforms hold credentials for many systems at once, so the security outcome depends on the deployment edition and the way those secrets are stored. Airbyte's documentation covers several arrangements, including open-source deployments, hosted offerings, external secret managers, and role-based access for supported environments. A control copied from one edition may not exist in another. Start with the deployed edition and version. Airbyte's security documentation for open-source version 0.44 describes a setup self-check that verifies internet accessibility and strong authentication. It also notes that the open-source product at that version did not include user management or role-based access controls. Newer or managed editions may expose organization and workspace roles. Verify the actual product surface instead of assuming that an RBAC page applies to every installation. The network check comes next. A connector service needs scheduled outbound access to source and destination systems, but its UI and API do not need to be public. Put the control plane behind an approved identity-aware proxy or private network, enable strong authentication, and test anonymous requests from outside the organization. If authentication is disabled because a proxy handles SSO, verify that the proxy covers every route, preserves secure headers, and does not allow cookie transmission over plain HTTP. Secret storage should be explicit. Airbyte's current architecture documentation says connector credentials are stored server-side and are not returned in API responses after creation. That behavior reduces one exposure path but does not make the database risk-free. The platform documentation recommends external secret management for production deployments. Confirm which secret-manager integration is active, who can change it, how keys rotate, and whether a failed secret lookup blocks connector execution. Privilege review should follow the data path. Organizations, workspaces, connectors, sources, destinations, and connections have different administrative actions. Administrators may be able to create a connection that writes to a sensitive target or reads from a broad source. Use the least-privileged role available in the deployed edition, separate production and development workspaces, and review service credentials for each connector. A connector that only reads one schema should not keep a database account with broader authority. API clients need lifecycle management too. Hosted Airbyte API credentials use client IDs and secrets tied to an organization. Store them in a secret manager, scope them to the required actions, and rotate them when integrations change. Audit new connector creation and configuration edits. A stolen API client can create a new data path even when existing connector credentials are inaccessible. A useful test creates a low-privilege user and a non-sensitive connector. Verify who can view the source configuration, edit the connection, trigger a sync, and read job logs. Confirm that logs and error messages do not reveal connector credentials or private endpoints. Then remove the user's role and repeat the checks after the documented propagation interval. Airbyte's exact security controls depend on edition and configuration. The review should therefore describe the deployment before claiming a protection. Authentication, secret storage, role boundaries, connector authority, and egress policy form the useful control model. Each one has an owner and a verification path. References - Airbyte platform security - Airbyte secrets management - Airbyte authentication configuration - Airbyte role-based access control

  3. Airbyte title matches are candidates until the control plane is confirmed Airbyte title matches are candidates until the control plane is confirmed A data-integration platform is a high-value control plane because a user can configure where credentials and records travel. Finding internet-visible Airbyte services helps an organization review ownership and authentication. A broad title match, however, is not proof of a confirmed Airbyte deployment and should never be presented as one. On 2026-10-28 at 21:41 UTC, a ZoomEye SDK probe used sub_type=all, page 1, and a page size of one. The query app="Airbyte" returned 1 match. The title query title="Airbyte" returned 12,861. The wide difference shows the limits of both observations. The application fingerprint may be narrow or incomplete, while the title search can include documentation, marketing references, and unrelated pages. Treat the title result as a candidate pool. Confirm the service through owner records or an approved response that identifies the product. Determine the deployment edition and version before making any claim about role-based access or secret handling. Airbyte documentation distinguishes open-source deployments, managed services, and supported RBAC features. A control present in one edition may not apply to another. Once an asset is confirmed, review the control plane as a credential delegation system. Ask which administrators can create sources, destinations, and connections. Review organization and workspace roles, API clients, authentication configuration, and external secret management. Test whether a low-privilege user can view connector settings or trigger a sync. Do not attempt to access credentials directly. Network exposure needs a clear reason. The service requires outbound access for connector jobs, but the management UI and API can remain private or behind SSO. Validate the public route, TLS, cookie settings, proxy coverage, and anonymous denial. If authentication is disabled behind a trusted proxy, request evidence that the proxy protects every path and does not forward credentials over insecure transport. Keep the evidence layers separate in the inventory. One field describes how the product was identified. Another lists the edition and owner. A third records authentication and network controls. Secret-manager integration and connector credential scope belong to a fourth review. ZoomEye fills part of the first field and can show change over time, but it cannot supply the rest. A repeatable query history avoids false precision. Report the app count, title candidate count, confirmed assets, and reviewed assets as separate measures. When the organization sees more public results, the next question should be ownership and control-plane access, not an automatic incident declaration. ZoomEye helps discover candidate data-integration services that may be missing from an internal catalogue. A confirmed Airbyte deployment then needs edition-specific authorization, secret storage, and egress review. The title match starts the investigation; it does not finish it. References - ZoomEye result for title="Airbyte" - Airbyte platform security - Airbyte secrets management - Airbyte authentication configuration

Achievements

Awards earned by the repositories airbytehq owns. Select an award to embed its badge.

No awards yet

Repositories earn awards by reaching the top three in a completed trending day or week.

A repository earns an award for finishing first, second or third in a completed trending day or week, ranked by new stars, new forks or pull request authors. Awards follow a repository through renames and transfers and count toward its current owner. Weeks run Monday to Sunday, UTC.