Comparisons¶
Each page says what Karni and the alternative both do, where each is the better choice, and who should pick which. Every page includes the rows where Karni loses: it is version 0.4, released in October 2026, with no known production deployment and no community yet. Facts about other projects were checked on 2026-10-21 and link to their sources.
- Karni vs cookiecutter-django: a generated Django project you own, against a framework you upgrade.
- Karni vs SaaS Pegasus: a commercial SaaS boilerplate with billing and teams, against an open-source framework without them.
- Karni vs plain Django + DRF: the standard stack, against the same stack with the per-model layers derived.
- Karni vs a Django React boilerplate: a template copied once, against a dependency that keeps receiving fixes.
In one table¶
| Plain Django + DRF | cookiecutter-django | SaaS Pegasus | React boilerplate | Karni | |
|---|---|---|---|---|---|
| You start with | an empty project | a generated project you own | a generated codebase you own (paid, with an open-source edition) | a copied repository | a dependency you upgrade |
| Per-model API, admin, permissions | you write them | you write them | generated once, then yours | you write them | derived from the model and access rule |
| An upstream fix reaches your code | for Django and DRF | no official path | regenerate and merge | no | karni upgrade |
| React front end | your own | not offered | HTMX or React | yes | @karni/ui |
| Subscriptions, teams, multi-tenancy | no | no | yes | some (Apptension) | no |
| Maturity | decades | since 2013 | since 2019 | since 2016 / 2020 | since October 2026 |
| AI agent guidance in the project | none | none in the generated project | rule files, skills, MCP | not checked | instruction files for eight tools, MCP, lint rules |