Compute
A test of 86 open-weight models shows large differences in energy use
Thinking mode uses about 3.7 times more energy on average. FP8 uses about one quarter less energy with a small change in score.
What matters in AI.
SubscribeNews category
71 stories, newest first.
Compute
Thinking mode uses about 3.7 times more energy on average. FP8 uses about one quarter less energy with a small change in score.
Compute
Makora charges $0.10 for 1M input tokens from cache, but Nebius charges $1.40 for all input, as Artificial Analysis found.
Compute
Four parts of the platform, with KAI Scheduler as one, can be used without the rest of the platform.
Compute
Meta, Microsoft and Alibaba operate a new group that sent its first specification to the Open Compute Project.
Compute
The Information reports that Amazon made Project Falcon, and the utility can put the data centers on backup power without contact with AWS.
Compute
One frontend with GraphTok processes 518 requests each second of 121,000-token traffic, against 296 with gigatoken.
arxiv.orgClaimed, not confirmed
Compute
AMD and TCS want to make a blueprint for a data center of 200 MW maximum in India.
Compute
The framework keeps 98.5% of the reference mAP and works on different GPU backends.
arxiv.orgClaimed, not confirmed
Compute
Omdia says a memory shortage from AI use increases prices and expects a 24% decrease in the fourth quarter.
Compute
Vegalabs says Microsoft and Anthropic each sent it to the other for a refund.
theregister.comClaimed, not confirmed
Compute
The authors write that the large model keeps 95 to 99% of its precision when it recomputes only 30% of positions.
arxiv.orgClaimed, not confirmed