Use Case: Comparing Visagism and Phenotype Facial Analyses Across Five AI Tools
1. Summary I uploaded profile photos of my face to five AI tools and asked each one the same questions about facial features, visagism impressions, and phenotyp
Latest DeepSeek — V4.1 with a Flash variant
“Who wouldn't be excited after finishing the test! I tested all night... even so, its performance was impressive enough.”
“Deepseek-v4-pro is 1.6T params, the cut-down version is 284B params, and I wouldn't trust them to do something on their own.”
“Over the past 24 hours, the consensus across the whole community of people who tried the GA v4-pro-0813 was that its felt capability fell far short of expectations, and third-party test scores could hardly match the numbers in the changelog.”
نعم — لدى DeepSeek V4 Pro درجة worth-it تبلغ 62%، بناءً على 318 صوتًا من المجتمع.
DeepSeek V4 Pro هو نموذج محادثة. Flash hits 300 tokens/sec on real tasks — feels instant
$0.66/$1.98 per 1M (off-peak)
1. Summary I uploaded profile photos of my face to five AI tools and asked each one the same questions about facial features, visagism impressions, and phenotyp
I Spent 400M Tokens Testing Kimi K3.1 — the Open-Source King Still Needs Work Last weekend Kimi K3.1 Pro dropped, and honestly I was hyped. The last K3 I touche
The receipts June–September Month Cost Tokens API requests --- --- --- --- June $112.78 3.68B 9,637 July $22.49 517M 2,069 August $15.11 155M 998 September $47.