Inkling 2 Mini Hits 210 tok/s Decode on Mac Studio, 11x Cheaper Than Sonnet
teortaxesTex · x · 2026-11-22
Exo Labs, working with Thinky Machines, made Inkling 2 Mini run fast locally:
- 96GB M5 Ultra Mac Studio: 210 tok/s decode, 3620 tok/s prefill
- 48GB M5 Max MacBook Pro: 125 tok/s decode, 1750 tok/s prefill
On the M5 Ultra it built a native weather app 1.8x faster than Claude Sonnet 5.5 and 11x cheaper at API pricing. The looping architecture suits memory-constrained local devices like MacBooks: looping layers raise inference-time compute (and intelligence) without growing model size.
More from Infra
- SpaceX lenders push back on $40B of AI chips launched into space — flngr · 2026-11-22
- Homie: an offline Mac notch voice assistant running Qwen 3.5 4B locally — KungFuSaifooo · 2026-11-22
- AMD and PyTorch ship a playbook to run GPT-OSS-20B inference locally with ROCm — PyTorch · 2026-11-22
- Microsoft's full Outlook outage PIR confirms AI tooling wasn't to blame — aronchick · 2026-11-22
- User reports 400+ tok/s with Flare inference engine — capetorch · 2026-11-22
- Huawei Ascend-950DT quoted at 3x 950PR price; yield and SuperPod interconnect may explain the gap — teortaxesTex · 2026-11-22