Waist-to-height ratio can help supplement the BMI measure without needing any special body fat scales or stuff like that.
It’s also a really simple one to remember. Keep your waist circumference below half your height.

Our aim was to differentiate the screening potential of waist-to-height ratio (WHtR) and waist circumference (WC) for adult cardiometabolic risk in people of different nationalities and to compare both with body mass index (BMI).
[…]
Compared with BMI, WC improved discrimination of adverse outcomes by 3% (P < 0.05) and WHtR improved discrimination by 4-5% over BMI (P < 0.01). Most importantly, statistical analysis of the within-study difference in AUC showed WHtR to be significantly better than WC for diabetes, hypertension, CVD and all outcomes (P < 0.005) in men and women. For the first time, robust statistical evidence from studies involving more than 300 000 adults in several ethnic groups, shows the superiority of WHtR over WC and BMI for detecting cardiometabolic risk factors in both sexes.

• https://www.researchgate.net/publication/51820603_Waist-to-height_ratio_is_a_better_screening_tool_than_waist_circumference_and_BMI_for_adult_cardiometabolic_risk_factors_Systematic_review_and_meta-analysis
  November 2011
  Obesity Reviews 13(3):275-86
  DOI:10.1111/j.1467-789X.2011.00952.x
  Source - PubMed

[…]the authors of the report determined that the energy used for a single query to the open-source Llama 3.1 8B engine used around 57 joules of energy to generate a response.[…]

A larger model, like Llama 3.1 405B, needs around 6,706 joules per response – eight seconds of microwave usage.

In other words, the size of a particular model plays a huge role in how much energy it uses.
Although its true size is a mystery, OpenAI’s GPT-4 is estimated to have well over a trillion parameters, meaning its per-query energy footprint is likely far higher than the Llama queries tested.[…]

AI video generation, on the other hand, is an energy sinkhole.

In order to generate a five-second long video at 16 frames per second, the CogVideoX AI video generation model consumes a whopping 3.4 million joules of energy – equivalent to running a microwave for an hour or riding 38 miles on an e-bike, Hugging Face researchers told the Tech Review.