Xybrid
PulseGate's liveness check found it on 8 Oct 2026; it is registered on GitHub and has been in the index since 7 Sep 2026. How this is checked
Xybrid is an inference layer for running machine-learning models on devices and routing requests to the cloud when local execution is unsuitable. It provides model loading, on-device defaults, cloud fallback, and SDKs for platforms including iOS, macOS, watchOS, tvOS, visionOS, Dart, Kotlin, C#, and Rust.
Inferred · not functionally tested
Overview
6 featuresPurpose: Running AI models across devices and cloud backends without managing infrastructure or sacrificing privacy and latency.
Inferred · not functionally tested
Audience: developers building cross-platform AI applications
Inferred · not functionally tested
Functions: Unknown
Interfaces: API: indicated (inferred, not tested) · MCP: unknown · CLI: unknown · Self-hosting: indicated (inferred, not tested)
Recorded constraints: pricing: unknown · license: Apache-2.0 · platforms: API, WEB · deployment: browser, embeddable, self_hosted, cloud_managed, api_only
Constraint provenance is unknown; confirm requirements with the publisher.
Record sources: xybrid.ai · github.com. These links do not verify the individual claims.
In the Inference & model serving space, Xybrid takes a focused approach. Inferred · not functionally tested: It focuses on running AI models across devices and cloud backends without managing infrastructure or sacrificing privacy and latency. Inferred · not functionally tested: Xybrid is a B2B product aimed at developers building cross-platform AI applications. Basis unknown · not verified: Xybrid is available on the web, embeddable surfaces, and API, and it can be self-hosted.
Behind Xybrid is Xybrid, and it first shipped in 2025. Development happens publicly on GitHub with 439 stars and 232 commits in the last 90 days. Inferred · not functionally tested: Among its 6 catalogued features are on-device inference, cloud fallback, and intelligent routing. Inferred · not functionally tested: Catalogued interfaces include a public API.
Summary written by a language model from the project’s public pages.
Tasks: Inferred · not functionally tested
- On-device inference
- Cloud fallback
- Intelligent routing
- Model registry
- Text generation
- Speech transcription
Topics: Inferred · not functionally tested
Built with & integrations
- Claude Code
- .claude/
- Vercel
- x-vercel-id header · x-vercel-cache header
- local_oss
- bllama in the HTML
- SvelteKit
- data-sveltekit- in the HTML · /_app/immutable/ in the HTML
- meta_llama
- llama- in the HTML
Trust & compliance
Indexing history
1What PulseGate has recorded for this listing
- Indexed7 Sep · 17:36 UTCxybrid-ai/xybrid seen via GitHub Search (Smart)Source: GitHub Search (Smart) · Open
Frequently asked questions about Xybrid
- What is Xybrid?
- Inferred · not functionally tested: Xybrid focuses on running AI models across devices and cloud backends without managing infrastructure or sacrificing privacy and latency. It is catalogued under Inference & model serving on PulseGate.
- Who is Xybrid for?
- Inferred · not functionally tested: Xybrid is a B2B product built for developers building cross-platform AI applications.
- What platforms does Xybrid run on?
- Basis unknown · not verified: Xybrid runs on the web, embeddable surfaces, and API. It can also be self-hosted.
- Is Xybrid still active?
- PulseGate's liveness check found it on 8 Oct 2026. Its GitHub repository shows 232 commits in the last 90 days.
- Who develops Xybrid?
- Xybrid is developed by Xybrid.
- How long has Xybrid been around?
- Xybrid first shipped in 2025.
- Is Xybrid open source?
- Basis unknown · not verified: Xybrid has a public GitHub repository.
- Does Xybrid have an API or integrations?
- Inferred · not functionally tested: Yes — Xybrid exposes a public API.
Similar projects
Closest matches by what these projects do