TensorFold 1.2.3 boosts Flash Next decode 21–23%; Apache-2.0 Ripple-65B Preview lands
AIFlow_ML · x · 2026-11-06
The 6 Nov 2026 Local AI Daily rounds up on-device/local deployment news:
- TensorFold 1.2.3 released; vendor-claimed +21–23% Flash Next decode speedup on 2× GB10 and a Mac parallel stall cut from 48s to 8s. Drop-in upgrade for existing Spark, Mac, Strix Halo, and PRO 6000 recipes on 1.2.2+.
- Ripple-65B Preview launched under Apache-2.0, a dense coding model sized to run on a single Spark.
- Also covered: a measured single-Spark GLM floor, mlx-serve Mac gains, and the llama.cpp v0.9.0 release.
Every figure is labelled by source type (MEASURED / VENDOR-CLAIM / ESTIMATE); window spans 4 Nov 00:30 to 6 Nov 07:15 Bangkok time.
More from Infra
- Zyphra ships ZYFA: FP8 attention on AMD MI355X at 1.16-1.76x AITER throughput — bclavie · 2026-11-06
- Akamai's signed AI book tops $14B, anchored by Anthropic's $11.6B commitment — pdamodaran · 2026-11-06
- Teknium pushes back on METR's 'no bottlenecks' claim, citing funding and compute cost inflation — Teknium · 2026-11-06
- Fudan's D3D decouples diffusion drafting for asynchronous speculative decoding speedups — Fudan-University · 2026-11-06
- ThursdAI pod: midterms, 40 datacenter ballot measures, $3T wipeout — is AI over? — altryne · 2026-11-06
- sshhole Bug Analysis: Alpine/MUSL Systems Hit Hard, ~100-500M OpenWRT Routers Were at Risk — esrtweet · 2026-11-06