OpenAI’s overall compute expenses

Data for OpenAI’s overall compute spending in 2024 come from Epoch’s AI Companies dataset, ultimately via reports in The Information and The New York Times. These reports indicate that OpenAI spent $3 billion on training compute, $1.8 billion on inference compute, and $1 billion on research compute amortized over “multiple years”. For the purpose of this visualization, we estimate that the amortization schedule for research compute was two years, for $2 billion in research compute expenses incurred in 2024. These are OpenAI’s cloud compute expenses in 2024, not the upfront capital cost of the data centers: OpenAI currently relies on cloud companies to access compute.

How exactly OpenAI distinguishes training and research compute is not known, so we group these expenses as $5 billion in total “R&D” (research and development) compute expenses.

Final training cost of released OpenAI models

For the training expenses of OpenAI models, we looked at OpenAI’s released or announced models from the end of Q1 2024 through the end of Q1 2025, as a rough heuristic to account for the delay between when a model was trained and its release/announcement date. Significant models from this time period include GPT-4.5, GPT-4o, o3 (December preview), and Sora Turbo.

We estimated the cost of the final training runs for these models, meaning the training runs that directly produced the released versions, as described below The code and full methodology can be found in this notebook.

GPT-4.5: We directly estimate GPT-4.5 cloud spending using estimates of GPT-4.5’s cluster size, duration, and price per GPU-hour, because we have more evidence about GPT-4.5 training cluster and duration than its training compute in terms of FLOP.

The training details we used were as follows:

Other models: We use a simpler approach of estimating their training compute in floating-point operations (FLOP), and then calculating cloud compute cost using standardized assumptions about GPU utilization and the cost of GPU-hours. We focus on confidence intervals of FLOP rather than point estimates in order to estimate the aggregate compute of OpenAI models. The confidence intervals we used are shown below:

These figures come from documents shared to investors in September 2024 and shown to journalists. They should not be considered completely reliable; for one, they rely on projections for the final quarter of 2024. Other lines of evidence suggest that these overall spending figures are approximately correct. OpenAI made around $3.7 billion in revenue in 2024, and is widely reported to be substantially unprofitable, meaning it spent much more than $4 billion in 2024. This would be consistent with ~$6 billion in compute spending, if compute makes up the bulk of their overall costs, and $1.8 billion in inference compute spending implies a plausible 50% margin on inference.

Note that the $1.8 billion in inference expenses presumably do not include Microsoft spending on inference to run licensed OpenAI models for its own products. Microsoft owns the data centers that OpenAI used for training, research, and inference in 2024, but Microsoft also separately licenses OpenAI models to power Microsoft products such as Copilot and the Azure API. These inference costs are likely significant but may be smaller than OpenAI’s own inference costs, since OpenAI’s products reportedly produce more revenue than their Microsoft equivalents.