See, understand and control your bills with dataintelligence.com

Your cloud and AI bills are growing and hard to understand. dataintelligence shows you exactly how to optimise your spend, eliminate waste and will never charge you more than you save.

Book your free 30-day diagnostic See the board

The board · tokensUSD per 1M tokens

Model	Input / 1M	Output / 1M	Cached input	Context	90 days
OA GPT-6 Astra OpenAI · flagship new 3 Sept	$10.00	$50.00	$1.00	1.05M	new	new 3 Sept
OA GPT-5.6 Sol OpenAI · upper-mid output $30.00 → $20.00	$4.00	$20.00	$0.40	1.05M	-33%	output $30.00 → $20.00
A Claude Fable 5.1 Anthropic · flagship cached $1.00 → $0.25	$10.00	$50.00	$0.25	200k	-75%	cached $1.00 → $0.25
A Claude Sonnet 5 Anthropic · mid held, rise to $3 / $15 cancelled	$2.00	$10.00	$0.20	200k	0%	held, rise to $3 / $15 cancelled
G Gemini 3.8 Flash Google · mid new 2 Sept, doubles 1 Jan 2027	$0.75	$3.75	$0.075	1.05M	new	new 2 Sept, doubles 1 Jan 2027
DS DeepSeek V4-Pro DeepSeek output $0.87 → $1.98	$0.66	$1.98	$0.022	1M	+128%	output $0.87 → $1.98

6 of 22 models changed a price in 90 daysThe full board

Essays

All essays

Dave Hackett30 September 2026

The cost of frontier AI work is rising

Imagine your team pays $200 a month for an AI plan. From today, a new subscriber on that plan gets half the usage for the same price. For new customers of OpenAI's Pro 200 plan, this is now true.

On 29 September, OpenAI launched a $500 a month Pro plan with 25 times the usage of ChatGPT Plus. At the same time, it cut the $200 Pro plan from 20 times to 10 times Plus usage. The $200 plan reopens to new subscribers today with the lower allowance. Existing subscribers keep their old allowance until 29 October.

Keep reading

Recent essays

An oil painting of a dartboard on a mustard wall, with a single dart stuck in a crack in the plaster well wide of the board.

Cheaper tokens, bigger bills. Why companies are moving to open-weight models

28 Sept · Dave Hackett

An oil painting of a one dollar bill burning at one corner on a mustard ground.

The Cloud and AI invoice grows every month, and nobody can quite explain why.

17 Aug · Dave Hackett

Connect, diagnose, optimise.

How it works in full

01 · Connect
Az
Azure cost export
14 months of invoices
read only
Az
Audit logs
who changed what, and when
read only
Az
Diagnostic files
what each machine is doing
read only
Sample data
We read the cost export Azure already writes. No agent on your machines and no write access.
02 · Diagnose
Six D8s_v5 machines below 4% CPU for 60 days
dev-east resource group · resize to D4s_v5
€2,140a month
38 E5 user licences with no sign-in for 30 days
remove or step down to E3
€2,050a month
1.8 TB of hot blob storage untouched since March
move to the cool tier
€310a month
8 findings · Sample data€12,750 / month
Within thirty days, a findings report. Every item carries what it costs you and how hard it is to fix.
03 · Optimise
A saving counts only when the next invoice shows it, and our fee comes out of that gap.