Which AI should you use? We test it and show the receipt.
Prices, limits and hands-on tests for the AI models and tools you are choosing between. Every number has a date and a source.
Latest test · Kingy measured
DeepSeek V4.1 Flash vs GPT-6, Claude and Kimi: 11 Models, 220 Trials + Cost
We reran the test on November 9, 2026 with 11 models. DeepSeek V4.1 Flash 1017 passed 20/20, but GPT-6 Luna also passed 20/20 at about a fifth of the cost per correct answer. Compare costs, failures and the test kit.
Kingy measured · receiptOriginal test
- Run
- Updated
- Sponsor
- No sponsor, Kingy paid
Latest
Claude Whiteboards, Sheets and Data challenge Microsoft 365
We asked 9 AI models 125 official figures. Here is how often they were wrong
Claude Fable 5.2 vs GPT-6.1 Sol: Is Fable worth four times the price?
Claude Fable 5.2 vs GPT-6 Astra: Benchmarks and task costs
How to get Fable 5.2: plans, limits, API pricing and apps
All news AI Price Index
NOV 2026AI prices held steady in November
Frontier model, per 1M tokens$4.00
Small model, per 1M tokens$0.26
Standard plan$20/mo
Top plan$200/mo
Changing · Gemini 4 Argon price doubles Dec 1
Open the indexFree tools
No account needed
Compare AI modelsTwo models side by side, from checked prices
Can you run it?Will a model fit your computer
Fix FinderPaste an Ollama, LM Studio or llama.cpp error
AI camera simulator16 camera moves for AI video prompts
Document to audioPDF, Word and EPUB to MP3
AI Certificate (Beta)Learn to use AI safely at work, then prove it
All tools The Kingy Brief
Every Friday: source-checked AI changes, an original test and one practical thing to try. Free, unsubscribe anytime.