Tag local AI

AI & Compute

Qwen 3.8 27B Needs Two RTX 5090s for Full Context

Three horizontal bars drawn to scale from one shared axis: two short gray bars for the 32,768 and 37,740 tokens of context a single RTX 5090 reached under the published vLLM and SGLang recipes, against a blue bar roughly eight times longer for Qwen 3.8 27B's full 262,144-token window.

Weights that fit in 32GB still left the RTX 5090 waiting half an hour for a first token, and the fix was a different inference engine rather than a different card.

Saya
September 9, 2026

Market & Pricing

The Memory Shortage Reaches the Top of the Stack: RTX PRO 6000 Now Costs $13,250

Two columns to true scale, launch price at 100 and today at 155, the 55 percent climb of the 96GB workstation flagship.

NVIDIA's 96GB workstation flagship has climbed 55% since launch, and reporting puts only a sliver of that on manufacturing. The rest is the same GDDR7 squeeze driving consumer prices, now visible at the very top of the product stack where there is no substitute to switch to.

Saya
July 30, 2026