Why Gated DeltaNet survives 4-bit: full NVFP4 W4A4 quantization of a hybrid 27B LLM
minima-ai · hf · 2026-09-04
minima-ai fully quantized a hybrid 27B LLM — including its recurrent Gated DeltaNet layers — to NVFP4 W4A4, preserving accuracy on long-context and reasoning benchmarks. The key is localizing outliers and exploiting the robust delta-rule dynamics, suggesting recurrent halves of hybrid LLMs are naturally low-bit friendly for edge deployment.
Related event: Minima Quantizes 27B Hybrid LLM Fully to NVFP4 W4A4(2 posts)→
More from Infra
- Samsung Electro-Mechanics joins Nvidia's substrate supply chain, shocking Taiwanese rivals — zephyr_z9 · 2026-10-23
- AI agents drove this dev's GitHub Actions bill from $738 to ~$2,400/month — CellistOk6782 · 2026-10-23
- Vercel founder Suhail publicly offering to buy RL kernel optimisation tasks for models — AIFlow_ML · 2026-10-23
- GitHub moved 99.99999% of repos to Azure; per-reference quorums unlock agent-scale Git concurrency — marlene_zw · 2026-10-23
- Jensen Huang visits Astana, joins Kazakhstan's AI minister to discuss nation's AI future — nvidia · 2026-10-23
- FP16 vs FP32 matmul on RTX 4060: 3.5x speedup at large sizes, slower at 128x128 — cneuralnetwork · 2026-10-23