Ramp data shows limited early business uptake for Anthropic’s Fable 5
Ramp’s August AI Index found that Fable 5 accounted for 6% of Anthropic token purchases and 11.4% of dollar spend in the month it measured. The company says the sample tracks its token-spend management users and skews more tech-oriented, so the figures are an early usage signal rather than a measure of all enterprise demand.
Ramp’s August AI Index reported that Anthropic’s Fable 5 made up 6% of the tokens businesses purchased from Anthropic over the month it measured, and 11.4% of dollars spent on Anthropic models. Ramp published the analysis on August 12, after Fable 5’s June release.
The result is a snapshot of purchasing behavior, not a direct test of model quality. Ramp says Fable 5 was its most expensive Anthropic model in the data and estimated its price at about $10 per million tokens. The company compared that early uptake with spending on other frontier models, arguing that stronger performance alone may not make the most expensive option the default choice for businesses.
What the index measured
Ramp says the figures come from its token-spend management product, which tracks daily token-use data at firms. It says the data are anonymized and aggregated, and notes that this Fable 5 sample is more technology-oriented than its usual AI Index sample. That makes the results useful for observing a particular set of business purchasers, but not a complete census of enterprise AI use.
Ramp also reported that 43.5% of U.S. businesses in its broader July index paid for Anthropic subscriptions or tokens. The Fable-specific figures therefore describe how a newly released model was used within Anthropic spending, rather than Anthropic’s overall share of the AI market.
Why the distinction matters
The Financial Times, in reporting republished by Yahoo Finance, highlighted the same early-spend pattern as a business-model question for frontier AI providers. The underlying Ramp figures support a narrower conclusion: in this measured cohort, the priciest model did not dominate Anthropic purchases during the period observed. They do not establish that all enterprises are rejecting Fable 5, nor do they predict its future demand.
For teams evaluating frontier models, the practical takeaway is to separate capability claims from workload economics. A useful internal comparison should measure task quality, latency and total token cost on the workloads a team actually runs, while treating third-party spend indexes as directional evidence with stated sampling limits.
Key Points
- 1Ramp’s August 12 index put Fable 5 at 6% of Anthropic token purchases and 11.4% of Anthropic-model dollar spend in its measured month.
- 2Ramp says the underlying token-spend sample is anonymized, aggregated and more technology-oriented than its usual AI Index sample.
- 3The figures describe early purchasing behavior in Ramp’s measured cohort; they do not measure model quality or all-enterprise demand.
Scoring Rationale
A first-party spend-data release offers a concrete, qualified view of early business adoption for a leading frontier model. The result is materially useful for teams comparing capability with token economics, while its single-provider sample limits justify a measured impact score.
Sources
Primary source and supporting public references used for this report.
Practice with real Ad Tech data
90 SQL & Python problems · 15 industry datasets
250 free problems · No credit card
See all Ad Tech problems
