Current view · observed 2 Jun – 29 Jun 2026 · 52 products · = Edition 1

The AI Credit Index.

What a credit actually costs across 50+ AI products — bundled rates, top-up premiums, and the expiry and exhaustion rules behind the meter, archived per row.

AI software is quietly migrating from seats to meters. Buy a plan and you no longer get features — you get a monthly allotment of credits, tokens, actions, or outcomes that deplete as you work. The AI Credit Index is a quarterly study of how that meter is priced and governed across 52 products, read only from vendors’ own public pages and archived row by row.

What you are reading This is the current view: the latest published observation for each product, observed 2 Jun – 29 Jun 2026. It is identical to Edition 1 (cut-off 2026-06-30) — nothing has been captured since. Cite an edition for a reference that will not move; download the JSON or CSV of this view; see how a row is built.

The headline number is a warning, not an answer: the median bundled rate is about 1.5¢ per credit, but published rates run from $0.0000018 to $25 — a spread of more than 13 million to one. A “credit” is not a unit; one buys a tweet in one product and a feature-film shot in another. So the Index never compares raw $/credit across products — it compares how products treat the same credit.

What a credit costs, by category
General agents (3) Incumbents adding meters (8) Media generation (16) GTM / RevOps (4) Content / marketing (3) Workflow automation (6) Coding / agents (9) Dev platforms (1) Support / CX (2) $0.000001$0.0001$0.01$1$100
category median (shown at n ≥ 5 products)

Each dot is one plan's bundled $/credit on a log scale (134 rates); aubergine marks each category's median, shown only where the category has ≥5 products. The shape is a map of unit-size: the cheap end meters tokens (coding, incumbents), the dear end prices whole outcomes or seats (support/CX, dev platforms).

SoF · The AI Credit Index
52 products with a published observation
203 plan tiers on those observations
+6% median top-up premiumn = 38 tiers with a public bundled rate and public pack price
40% of top-ups hide the priceof 139 tiers that sell a top-up

Read this three ways

The Index is built in three layers. Take the facts as a citable source; use the analysis to see the patterns; then turn it into your own pricing decisions.

Or jump to Packaging (free tiers, grants, annual terms), The fine print (expiry, overage, what happens at zero), Transparency, or What’s changing. For a single product’s history — its rung, its moves since January 2025, and its likely next step — see the per-product Reviewed pages. The full panel, source ledger, and method live under Methodology & data.

Direct answers

What does an AI credit cost?
Edition 1, observed 2 Jun – 29 Jun 2026: across 52 AI products with public credit pricing, the median bundled rate is about 1.5¢ per credit ($0.01539), computed over 134 paid plan tiers that publish both a price and an included allotment. The median is close to meaningless on its own: published rates span $0.0000018 to $25 — more than 13 million to one — because a “credit” buys a different amount of work in each product. The Index compares how a product prices against itself, not raw $/credit across products. Full table →
How common are AI top-up premiums?
Edition 1, observed 2 Jun – 29 Jun 2026: of the 38 plan tiers where both the bundled rate and the top-up price are public, 20 charge more per credit to top up than the plan itself, 14 charge less, and 4 charge the same. The median top-up premium is +6%; the highest is +350%. Plans whose top-up price is hidden (56 of 139) cannot enter this figure, so the true premium across the market is unknown, not zero. Full table →
Which AI credits expire, and which roll over?
Edition 1, observed 2 Jun – 29 Jun 2026: included credits expire at the end of the period on 116 of 203 plan tiers and roll over on 40; 46 tiers do not say. Purchased top-up credits are treated more generously: they roll over on 7 tiers and expire on 53, with 112 unstated. Unstated is recorded as unstated — never assumed either way. Full table →
What happens when users exhaust their credits?
Edition 1, observed 2 Jun – 29 Jun 2026: the most common answer across 203 plan tiers is a hard stop — 111 tiers block further use until the next reset or a purchase. 30 tiers offer overage you opt into, 21 bill overage automatically by default, 15 throttle rather than stop, and 26 do not state what happens at zero. Full table →
Which AI companies disclose top-up pricing?
Edition 1, observed 2 Jun – 29 Jun 2026: of the 139 plan tiers that sell a top-up, 60% publish the price on a pricing page (36) or in help docs (47); 40% show it only at the in-app purchase screen (48) or not at all (8). The Transparency tab lists every product by disclosure; a further 13 vertical-AI leaders publish no pricing page at all and sit on the opacity watchlist. Full table →
How do AI companies package credits?
Edition 1, observed 2 Jun – 29 Jun 2026: 26 of 52 products open with a free credit grant, refilled daily on 6, monthly on 16, and never (a one-time sampler) on 3. 12 products (23%) peg one credit to a fixed dollar value; 23 vary how many credits an action burns by the model used. Annual billing, seat scaling and trial grants are tabulated on the Packaging tab. Full table →
How do AI coding tools price credits?
Edition 1, observed 2 Jun – 29 Jun 2026: the 9 coding-agent products in the panel meter at the cheap end of the market — median bundled rate $0.38333 per credit over 18 paid tiers, against $0.01539 for the whole panel — because the unit is a token-sized action rather than a finished output. Where the top-up price is public (6 tiers) the median premium is −3%. 22 of the 28 coding tiers with a top-up hide its price. Full table →
How do consumer and business AI credit models differ?
Edition 1, observed 2 Jun – 29 Jun 2026: the panel is tagged by primary audience — 4 consumer, 24 prosumer, 24 business products. The consumer group is below the Index’s five-product floor, so no consumer aggregate is published. Among prosumer products 62% of 71 tiers with a top-up hide its price, against 70% of 57 for business products; median bundled rates are $0.005 (prosumer) and $0.04 (business). The Big Picture tab plots audience against the opacity composite. Full table →
How is the AI Credit Index sourced and verified?
Every figure comes from a vendor’s own public pricing page or help docs — never third-party trackers, leaks or estimates. Each published product is captured to a dated observation backed by at least one Wayback Machine snapshot, and every derived number is computed from those observations at build time, never hand-entered. Where a price is hidden behind the logged-in purchase screen it is recorded as hidden, not guessed: the disclosure rate is one of the things measured. Aggregates are suppressed below 5 products; nothing is imputed. Full table →

Cite this research

Publisher
Strategy of Finance (Rohit Agarwal)
Title
The AI Credit Index (current view)
Version
current — moves when a capture lands
Data window
observed 2 Jun – 29 Jun 2026
Page generated
Sample
52 products · 203 plan tiers · 134 bundled rates · 38 top-up premiums
Canonical URL
https://www.strategyoffinance.com/benchmarks/ai-credit-index/
Licence
CC BY 4.0 — reuse with attribution to “Strategy of Finance — The AI Credit Index”. Reuse terms.

For a reference that will not move, cite the edition: Edition 1 — identical to this view today.

Suggested citation

Strategy of Finance (2026). The AI Credit Index (current view) — data observed 2 Jun – 29 Jun 2026; page generated 21 September 2026. https://www.strategyoffinance.com/benchmarks/ai-credit-index/
BibTeX 
@misc{sof_ai_credit_index_current,
  author       = {{Strategy of Finance}},
  title        = {The AI Credit Index (current view)},
  year         = {2026},
  month        = {jun},
  howpublished = {\url{https://www.strategyoffinance.com/benchmarks/ai-credit-index/}},
  note         = {Current view; data observed 2026-06-02 to 2026-06-29; accessed 2026-09-21. Licence CC BY 4.0.}
}
JSON CSV Methodology Field definitions Change log