It appears on the home page and opens for votes when its launch week starts.
YABANIN
AI adoption. Your infrastructure. Measurable value.
FreeOpen SourceAIAnalytics5 impressionslaunches Nov 171 comment
Comments
>log in to comment
We built YABANIN to give companies a self‑hosted gateway that lets them control AI model access, routing, budgets, and usage analytics on their own infrastructure. It lets you pick any model, set cost limits, and see who’s using AI and at what cost. Please try the budget limits feature and let us know if anything breaks.
More Open Source tools →74 free dev tools →
- Self‑hosted AI gateway with routing, budgets and analytics
- Privacy‑first design – no cookies, no session recording
- RAIDER: repeatable Scala agent workflows for engineering
- Cost‑by‑task and team analytics with real‑time usage tracking
- Open‑source Apache‑2.0 core, enterprise deployment options
Self-hosted AI gateway that lets enterprises control model access, routing, budgets and usage analytics.
- for
- Engineering and ops teams in mid-size companies that need on-prem AI control.
- pricing
- contact sales
- license
- Apache-2.0
- v0.1.0-preview.1Yabanin Core v0.1.0-preview.13 days ago
Key features
- Self-hosted gateway — Runs in your environment to expose approved model routes and enforce policies.
- Access control — Scoped credentials and approved routes limit who can call which model.
- Routing — Unified gateway routes requests to multiple model providers.
- Budgets — Set spend limits per team, task or agent and enforce them automatically.
- Usage analytics — Record who used AI, for what, and at what cost for each call.
- Agent workflows (RAIDER) — Compose repeatable engineering workflows in Scala that run locally or in CI.
Use cases
- Enforce data-boundary policies by keeping inference on-prem while allowing approved external models.
- Attribute AI usage cost to specific tasks, agents or teams to evaluate ROI.
- Create repeatable code-review or CI automation agents that call models through a controlled gateway.
- Test alternative models against quality requirements before changing production workflows.
YABANIN vs alternatives
 YABANIN |  OpenRouter | .png?width=64) Langfuse |  PostHog |  Mixpanel | |
|---|---|---|---|---|---|
| Best for | Self-hosted AI access with cost visibility | Hosted multi-model API | LLM observability | Product analytics | Event analytics |
| Pricing | Contact sales | Subscription | Subscription | Subscription | Subscription |
| DevHunt upvotes | 1 | 0 | 89 | 0 | 0 |
| Launched | Nov 2026 | — | Jan 2023 | — | — |
- YABANIN vs OpenRouter: Provides a single API for many models but is a hosted service, not self-hosted with budget controls.
- YABANIN vs Langfuse: Focuses on observability for LLM apps, not on routing, budgets or self-hosting.
- YABANIN vs PostHog: Product analytics platform; measures usage but does not control model access or costs.
- YABANIN vs Mixpanel: Analytics for events and funnels; no AI gateway or budget enforcement features.
YABANIN FAQ
How does YABANIN keep my data private?+
The platform runs as a self-hosted gateway in your infrastructure, so data never leaves your environment unless you explicitly route it to an approved provider.
Can I limit how much AI a team can spend?+
Yes, YABANIN lets you set budget limits per team, task or agent and enforces them at request time.
What models can I use with YABANIN?+
You can connect any approved model provider through the gateway or run local models you host yourself.
What is RAIDER?+
RAIDER is YABANIN’s agent workflow engine that lets you compose engineering tasks in Scala and run them locally or in CI with a full audit record.
Is there an open-source component?+
The core AI-infrastructure is Apache-2.0 licensed and available as source on request; enterprise deployment is provided by the YABANIN team.
Summarized by DevHunt from yabanin.ai · Nov 11, 2026. Details may change; check the official site.
