The open-source solution to building, maintaining, and collaborating on GraphQL Federation at Scale. The alternative to Apollo Studio and GraphOS.
1.2k
Stars
242
Forks
120
Open issues
30
Contributors
AI Analysis
WunderGraph Cosmo is an open-source GraphQL Federation platform that provides schema registry, composition checks, analytics, metrics, tracing, and routing capabilities as an alternative to Apollo Studio and GraphOS. It is purpose-built for organizations managing federated GraphQL architectures at scale, serving platform engineers, architects, and developers who need to coordinate multiple GraphQL services into a unified graph; it is not a general-purpose GraphQL tool but rather a specialized...
Inferred from signals mentioned in the README (tests, CI, type safety) — not a review of the actual code.
AI's overall editorial judgment — not an average of the bars above, can weigh other factors too.
Open-source GraphQL Federation platform with managed and self-hosted options, positioned against Apollo's commercial offering
Cosmo is an Apache 2.0 GraphQL Federation platform encompassing schema registry, composition, analytics, metrics, and routing. Built by WunderGraph, it targets teams operating federated GraphQL architectures who want to avoid Apollo vendor lock-in or need on-premises deployment. The project bundles infrastructure typically sold as SaaS (Apollo Studio/GraphOS) into an open, deployable system. Real-world adoption appears concentrated in organizations already familiar with WunderGraph or those evaluating federation at scale.
Created August 2023 by WunderGraph (a company providing GraphQL development tools). Cosmo emerged as WunderGraph's answer to the market gap between open-source federation composition (Apollo Federation) and proprietary managed services. It represents a deliberate positioning against Apollo's commercial platform rather than organic evolution of a community project.
Gained ~1,200 stars over ~3 years, adding ~7 stars in the last 7 days (2026-07-06). Growth is slow and linear rather than viral. README references 'State of GraphQL Federation 2024' report and production deployments (e.g., 'SuperBowl scaling'), suggesting deliberate enterprise positioning. The growth pattern indicates niche adoption among federation practitioners rather than broad ecosystem traction. WunderGraph's active development (last push same day as analysis date) shows corporate commitment but not explosive community momentum.
README cites scaling experiences (SuperBowl deployment, observability architecture blog posts), suggesting production use at some scale. Blog posts reference technical optimizations (query planning, subscriptions, dataloader) that imply real workloads. However, no public case studies, customer count, or deployment statistics provided. Adoption not systematically verified but blog content suggests real-world validation beyond toy examples.
Appears to be a modular TypeScript/Go stack with: router (Rust/Go-based for performance), schema composition engine, analytics/observability layer, and schema registry. README indicates support for Kubernetes deployment via Helm, Docker Compose for local testing, and managed cloud service option. Router implementation leverages systems programming languages (Rust/Go) for high-throughput federation workloads. CLI and server components decoupled for flexibility. Specific implementation details not verifiable from README alone.
Not documented in README. No mention of test suites, CI/CD practices, or coverage metrics.
Active maintenance: last push 2026-07-06 (same day as analysis). Repository shows regular activity with examples for different personas (developer, architect, platform engineer). Contributing guide present and changelog maintained. No indicators of abandonment or slow releases. However, modest GitHub activity (240 forks, 1,239 stars, 7 stars/week) suggests a maintained but not high-velocity project.
ADOPT IF: your team operates federated GraphQL at scale and requires either on-premises deployment or wants to avoid Apollo commercial pricing/lock-in; you have the platform engineering capacity to maintain a multi-component system or willingness to use the managed cloud service. AVOID IF: you are just starting with GraphQL and federation adds unnecessary complexity; you need a lightweight, single-binary solution; you require 24/7 commercial SLA support from an established vendor. MONITOR IF: you are evaluating federation architecture now and want to track how Cosmo's adoption and feature parity evolve relative to Apollo over the next 12–18 months; you are considering WunderGraph's ecosystem more broadly.
Independent dimensions
Mainstream potential
4/10
Technical importance
7/10
Adoption evidence
5/10
- Adoption appears concentrated in WunderGraph ecosystem / brand awareness; may struggle to grow beyond that installed base without significant marketing or partnership efforts.
- Competitive pressure from Apollo's continued investment in GraphOS and federation tooling; Apollo's 13,900+ star lead reflects market dominance and community network effects.
- Operational burden on teams adopting Cosmo for on-premises use; requires Kubernetes or Docker expertise, schema registry operations, and routing layer management.
- Incomplete feature parity with Apollo Studio (e.g., schema checks, composition validation, cost analysis) not explicitly detailed in README; feature comparison opaque.
- Dependency on WunderGraph's continued corporate commitment; while open-source, the project appears tightly coupled to WunderGraph's commercial vision and may not sustain if company priorities shift.
Cosmo likely remains a niche but sustainable project, valuable for organizations committed to federation and on-premises requirements or Apollo cost avoidance. Mainstream adoption unlikely unless Apollo raises prices significantly, WunderGraph secures major enterprise customers, or the community dramatically expands contribution. Most probable trajectory: slow, steady growth to 2,500–5,000 stars by 2028; establishes itself as the credible open-source alternative to Apollo for specific use cases rather than a general-purpose replacement.
Newsletter
Get analyses like this every Monday
Free weekly digest of the most interesting open-source discoveries.
Languages
Information
- Language
- TypeScript
- License
- Apache-2.0
- Last updated
- 18h ago
- Created
- 35mo ago
- Analyzed with
- anthropic/claude-haiku-4-5
Stars over time
Contributors over time
Top 100 contributors only — repos with more will plateau at 100.
Top contributors
Similar repos
ardatan/graphql-mesh
GraphQL Mesh is a federation framework and gateway that converts non-GraphQL...
APIs-guru/graphql-voyager
GraphQL Voyager is an interactive visualization tool that renders any GraphQL...
apollographql/apollo-server
Apollo Server is a production-ready, spec-compliant GraphQL server for Node.js...
| Repository | Stars | Week Δ | Language | Score | Updated |
|---|---|---|---|---|---|
|
wundergraph/cosmo
this
|
1.2k | +7 | TypeScript | 8/10 | 18h ago |
 cosmosgl/graph
|
1.2k | — | TypeScript | 8/10 | 17h ago |
|
ardatan/graphql-mesh
|
3.5k | — | TypeScript | 8/10 | 19h ago |
|
APIs-guru/graphql-voyager
|
8.2k | — | TypeScript | 8/10 | 2mo ago |
|
apollographql/apollo-server
|
13.9k | — | TypeScript | 9/10 | 16h ago |
|
ardatan/graphql-tools
|
5.4k | — | TypeScript | 8/10 | 2d ago |
Apollo Server dominates open-source federation; Apollo Studio/GraphOS provides the managed platform Cosmo mimics. Cosmo's key difference: open-source + on-prem deployment without vendor lock-in. Apollo Server remains necessary for subgraph development; Cosmo competes at the federation platform/gateway layer, not server level.
Mesh is a broader API composition tool covering REST, GraphQL, SOAP, and other protocols. Cosmo is Federation-specific and deeper in that domain. Different use cases; Mesh for API aggregation, Cosmo for federated GraphQL governance.
Cosmo assumes teams already operating federation. Monolithic approaches (single resolvers per schema) sidestep federation complexity entirely. Cosmo is not a replacement; it's a tool for organizations that have already chosen federation as their architecture.
Teams can build federation platforms manually using Apollo's composition library + custom routing. Cosmo packages this as a turnkey offering with observability, schema registry, and routing built-in. Reduces engineering effort but sacrifices bespoke optimization.
Hasura provides instant GraphQL over databases and APIs, federation-agnostic. Targets different personas (rapid prototyping vs. structured federation governance). Not direct competitors; different problem spaces.
