GitHub Repository

On-device AI SDK for Flutter — LLM inference, vision, STT, TTS, image generation, embeddings, RAG, and function calling. Metal GPU on iOS/macOS.

84 starsDart

Edge Veda – A framework for resource-aware edge computing

Name: Edge Veda – A framework for resource-aware edge computing
Availability: InStock
Author: ram2497

by ram2497·Feb 19, 2026·3 points·4 comments

AI Analysis

●●●BangerWizardryBig BrainSolve My Problem

Thermal-aware runtime keeps mobile models stable 28min vs 2min crashing.

Strengths

•Solves genuine mobile AI fragility: thermal throttling, memory spikes, session death after ~60s—not just a wrapper
•Persistent worker model with supervised resource policies is architecturally sound; not just inference calls
•Benchmarked degradation curves (with screenshots) show engineering rigor missing from most ML mobile tools

Weaknesses

•iOS-only currently limits reach despite strong execution; Android roadmap unclear
•RAG + speech require model downloads; documentation light on typical model sizes and storage costs