Toot blog
8 July 2026
Blog
Notes from building an AI bookkeeper: benchmarks, engineering, and what AI can — and can't yet — do with real company books.
GLM 5.2 is (nearly) as accurate as a human book-keeper at less than 1% of the cost
We benchmarked an open-weights AI model on a real quarter of UK SME books: 59 transactions, 68 minutes, $2.73 of tokens — and a VAT return within 7 pence of the human-prepared one. What it got right, what it got wrong, and why it matters.
Read the post →