Credits as the universal currency
123 of 338 corpus companies meter usage in an internal 'credit' currency that floats across features -- still the #1 *consumption* meter by a wide margin, but seats (175) now lead the corpus overall by a clear gap. Credits are the standard architecture in AI-coding and outbound sales-tech, beyond the creative-app origin; the abstraction is a usage-layer phenomenon, not the corpus-wide modal unit. Simpler invoice, weaker transparency.
What's happening — and why
What's happening: instead of charging separately for each feature — a minute of video, a thousand tokens, one image — many vendors sell a pool of generic 'credits' that every feature draws from at its own rate. Credits are the most common *consumption* meter in the corpus (123 of 338 companies), far ahead of tokens (66) and other usage units. But they are not the modal unit overall: seats appear in 175 companies, because credits live on the usage layer while seats anchor the base of seat+credit hybrids.
Why: when a single product spans wildly different cost structures, one currency makes the plan easy to present and lets the vendor adjust the dollar-to-compute ratio without republishing prices. The trade-off is transparency: buyers can no longer see what any given action actually costs, and the conversion rate can move under them. As the corpus broadened to 338 companies the seat-vs-credit gap widened — credits meter what you do, seats meter who does it, and most enterprise SaaS still bills the seat first.
How it works
Evidence over time
15 supporting · 2 counter — hover or tap a point for detail, click to jump to the row.
Evidence
| Company | Date | What happened |
|---|---|---|
| GitHub Copilot | Jun 2026 | AI Credits (1 credit = $0.01 USD) replaced premium requests across individual and org plans — a credit currency reaching a developer-tools enterprise buyer. |
| HeyGen | May 2026 | Cut over to a unified credit-based model; Generative Credits renamed Premium Credits (2026-05-04), one currency across avatars, dubbing and video. |
| Cursor (Anysphere) | Jun 2025 | Switched from request-based billing to a credit-pool model (the change that later triggered the July 2025 refund apology). |
| Recraft | Dec 2024 | Credit-pool overhaul replaced the flat $20 tier with Free/Basic/Advanced/Pro credit buckets after Recraft V3. |
| Descript | Sep 2025 | Media minutes + AI credits replaced transcription-hours as the metered unit. |
| Synthesia | Jan 2025 | Unified a single credit pool across all AI features. |
| Codeium | Jan 2026 | Windsurf Pro repriced at $15/mo with an expanded credit pool as the agentic-coding meter. |
| Cartesia | Sep 2024 | Tiered subscription plans denominated in credits with usage overages. |
| Ideogram | May 2026 | Free/Plus/Pro/Team tiers all metered in credits across image generation. |
| Lovable | Jul 2025 | App-builder platform metered entirely in credits — Free (5 msgs/day), Starter $20/mo, Launch $50, Scale $200 — a single credit unit across all app-building actions. |
| Creatify | Jun 2026 | Video-ad platform with credits + seats + media-minutes — three-unit structure, but credits are the primary consumption unit. |
| Bardeen | Jun 2026 | Automation tool metered in credits (actions) — free tier with credit allotment, paid tiers add credits — the credits-for-automations pattern mirrors outbound-sales tooling. |
| Qodo | Jun 2026 | Migrated from per-seat Teams to a pure team-wide credit pool ($.012/credit, sold in 2.5K/5K/20K packs) — credits replacing the seat base entirely, the strongest form of credit abstraction in the corpus's AI-coding segment. |
| Dust | Jun 2026 | Converted a flat €29/seat unlimited plan into credit-bundled seats (Pro 8,000 credits/mo, Max 40,000; Basic 1 / Advanced 3 credits; $0.01/credit programmatic) — unlimited usage re-expressed as a floating credit currency. |
| Zapier | Jun 2026 | Unified AI steps, code, MCP and the SDK into a single shared task pool with variable per-action rates (rates vary by AI model tier, code runtime, connector type) — a task is now a complexity-weighted credit, converging the automation meter onto one abstracted currency. |
Trivia
-
Credits became the most common billing unit in the corpus (roughly 35% of companies with taxonomy data at 97 companies), yet no two credit systems define a credit the same way: Cursor's $1 credit = $1 of underlying API cost is the most transparent, while most creative-tool vendors deliberately decouple the credit from any published dollar-to-compute ratio so they can change underlying costs without renegotiating. The same word describes a spectrum of disclosure from fully transparent to fully opaque.
-
The 71% correlation between credit-metered vendors and freemium tiers (observed at 61 companies) is the strongest cross-axis correlation in the corpus's credit analysis — stronger than the credit-enterprise or credit-hybrid correlations. The credit pool is architecturally suited to freemium because the free monthly allotment and the paid pool share one currency, making the upgrade trigger (running out of credits) unambiguous to the user.
-
The credit-currency pattern spread into enterprise software via GitHub Copilot's June 2026 AI Credits launch — the first time a product with tens of millions of users adopted the credit abstraction for its primary metering unit. Prior to that, the pattern had been concentrated in creative-app and developer-tool SMB segments; GitHub's adoption confirmed that the credit abstraction scales to enterprise procurement volumes.
-
At 158 companies the credit lost its crown: seats (78 arrays) overtook credits (70) for the first time across four syntheses, after credits had led at 122 (59) and every prior count. The reversal wasn't a flight from credits — it was the corpus absorbing 36 new companies skewed toward seat-anchored vertical SaaS, exposing that "credits are the most common unit" was always a sampling artifact of an AI-native-heavy corpus.
-
Credits and seats turned out not to compete but to stack: in the seat+credit hybrids that dominate AI-coding and outbound sales-tech, the same bill carries both units, so a company can push both counts up at once. That's why seats (78) and credits (70) can each be "near-majority" simultaneously in a 158-company corpus — they measure orthogonal things (who logs in vs. what they consume), not rival pricing philosophies.
-
Qodo's 2026-06-30 switch breaks the "credits and seats stack, never substitute" finding: it dropped per-seat pricing entirely for a pure credit pool, the corpus's first AI-coding vendor where the credit didn't layer on top of the seat but replaced it. The orthogonality holds as a corpus-wide average, but at the vendor level the credit can now eat the seat — when the team-wide pool is the only meter, "who logs in" stops being billable at all.
For buyers
A credit hides the dollar-to-compute conversion, and that rate can change unilaterally — Cursor's 2025 credit-pool switch was opaque enough to force a public refund. Before committing, ask for the dollar value of a credit per feature and whether that ratio is contractually fixed. Note where the credit sits: in most plans it rides on top of a seat (you pay for both), but a handful of vendors (Qodo's 2026 switch to a pure team-wide pool) now let the credit replace the seat entirely.
For vendors
A credit economy needs a metering layer that converts every feature's real cost into credits, a balance/top-up system, and clear in-product disclosure of burn rate — plus the governance to change conversion rates without a trust rupture. Credits remain the natural meter for the usage layer; for the base, seats still dominate the corpus (175 vs 123), so most vendors stack credits on a seat rather than replace it.
Outlook — what to watch
Credits will keep spreading on the usage layer as vendors ship more heterogeneous features (video, voice, agents) under one plan — where vendors restructured usage in 2026 (Qodo, Dust, Zapier), the credit or task pool was the destination. But the corpus-wide lead stays with seats, and that gap has widened, not closed: the thesis weakened from 'most common unit' to 'most common consumption unit.' The status flips back toward 'holds/sharpens' only if seat-anchored vendors begin dissolving the seat into the pool the way Qodo did; it weakens further if more vendors reprice without credits at all (Shortwave, Creatify, lemlist all did in 2026).
Bottom line
123 of 338 corpus companies meter in credits — the most common consumption unit by a wide margin — but seats (175) clearly lead the corpus overall, and the gap has widened. Credits simplify the invoice and decouple price from compute on the usage layer, at the cost of unit-economics transparency; they are not the corpus-wide modal unit.
FAQ
What are AI 'credits' and why do companies use them?
A credit is a synthetic currency that converts the cost of many different features (a minute of video, a thousand tokens, an agent run) into one unit. It simplifies the bill and lets vendors change the dollar-to-compute ratio without restating prices.
Are credits the most common AI billing unit?
They're the most common *consumption* unit — 123 of 338 corpus companies meter in credits, far ahead of tokens (66). But they aren't the most common unit overall: seats lead at 175, because credits ride on the usage layer while seats anchor the base of seat+credit hybrids. Credits meter what you do; seats meter who does it.
Are credit-based plans good or bad for buyers?
Convenient but less transparent. Because the conversion rate is set by the vendor and can change, it's harder to model true unit cost. Ask for the dollar value of a credit per feature and whether it's contractually fixed.
Which AI companies use credit pricing?
Across the corpus, 123 of 338 — including HeyGen, Cursor, Recraft, Synthesia, Descript, Cartesia, Ideogram and Qodo, plus dev tools (GitHub AI Credits) and enterprise search (Glean FlexCredits). Pure-token API players like Anthropic and Cohere deliberately don't.