1/ Daloy cannot automatically enforce branch protection, environment approvals, secret hygiene, runner isolation, egress policy, org settings, or deploy-platform configuration in a user’s private repo.

Hono: smallest portable router. Fastify: mature Node ecosystem. Elysia: Bun-first TypeScript magic. Daloy: API contracts, docs, generated clients, tests, runtime portability, and secure defaults from one source of truth.

Using @nextjs and deploying to @vercel automatically results in a 100 Lighthouse score without doing anything. 🤯

We built a fake rest api with 600 endpoints using our own web framework with @scalar docs, deployed in @vercel. If you want to try the latest @expo's SDK 56, use this as your backend for free. Don't forget to use @mrlubos' Hey API + @tan_stack query 🙂. fakerestapi.vercel.app/docs

Smaller bundles, same contract. Use @valibot anywhere DaloyJS expects a Standard Schema — including edge deployments where every kilobyte matters. daloyjs.dev/docs/validation/…

The @zodtypes cookbook for DaloyJS: request bodies, query params, headers, response schemas, discriminated unions, and how it all lands in your OpenAPI doc. daloyjs.dev/docs/validation/…

Standard Schema in, validated input out. DaloyJS works with Zod, Valibot, and anything that speaks the spec — no adapters, no wrappers. daloyjs.dev/docs/validation

`pnpm gen` turns your OpenAPI doc into a fully typed fetch SDK. Zero config, no extra code, and your frontend imports it like any other package. Because we love Hey Api @mrlubos ❤️ daloyjs.dev/docs/typed-clien…

DaloyJS is a runtime-portable TypeScript framework in which a single route definition provides validation, types, OpenAPI, typed clients, and contract tests. Start with the big picture here. daloyjs.dev/docs

Qwen 3.7-max beats Opus 4.7 and GPT-5.5 We tested three frontier models on a real agentic task: write a Tetris bot that plays the game and trains itself. Each model could read its own code, run benchmarks, and rewrite itself across 10 iterations. Then we compared the final bots head to head. Qwen 3.7-Max: training cost $1.32, bot improvement +56% Claude Opus 4.7: training cost $12.15, bot improvement +28% GPT-5.5: training cost $2.85, bot improvement +7% Qwen won on every dimension - biggest jump, 9× cheaper than Claude, 2× cheaper than GPT. Long agentic loops is where Qwen Max actually delivers.

Our new documentation is built using @nextjs, @shadcn, and @tailwindcss. daloyjs.dev/

contract-first + typed client codegen + security-by-default 🙌 daloyjs.dev/ on public preview.