DART Enables Real-Time SAM3 Detection with INT8 Quantization for Edge
rsasaki0109 · x · 2026-08-29
Mehmet Kerem Turkcan released DART, a training-free framework converting SAM3 into a real-time multi-class open-vocabulary detector. It achieves 15.8 FPS on an RTX 4080. The upgrade, DARTF, uses rotation-based quantization to enable INT8 inference for LayerNorm vision transformers. On Jetson AGX Orin, DARTF is 43% faster and uses 42% less energy than DART's FP16 engine while maintaining FP32-level quality (56.0 AP).
Related event: DART Turns SAM into a Real-Time Open-Vocabulary Detector Without Training(2 posts)→
More from Infra
- Nvidia's developer base hits 11 million, up ~3.7x from 3 million in 2021 — Beth_Kindig · 2026-10-16
- Muse Spark Open runs locally on a Mac M3 Ultra via llama.cpp: frontier model on desktop — pcuenq · 2026-10-16
- Muse-Spark open-weights drop: 816B-A42B MoE model gets 359GB MLX 3-bit build — ns123abc · 2026-10-16
- Meta launches Muse Spark Open, says it runs on two M5 Ultra Mac Studios with Apple's help — The Verge AI · 2026-10-16
- vLLM ships Day-0 LoRA RL for Muse Spark Open, lifting AIME from 0.56 to 0.85 — vllm_project · 2026-10-16
- DSpark draft model for Muse Spark Open proposes 7 tokens per pass with 23-42% acceptance — vllm_project · 2026-10-16