Model download statistics

[Ph94]

I’m working on an academic project on how users select models when they have increasingly more options. For this, I’m collecting daily data on model downloads on Hugging Face. I, however, noticed that the total number of downloads decreases for some models between days. For example, the picture below shows it for OpenAI’s Whisper small model between 8/3/2025 and 9/3/2025.

Could someone explain why this is the case?

To collect the data, I’m running:

model_list = list(api.list_models())

I run that code daily at midnight.

Thanks in advance!

[John6666]

I think this is because it’s not the total amount of downloads, but the number of downloads in the last 30 days.

HfApi Client

• downloads (int) — Number of downloads of the model over the last 30 days.
  downloads_all_time (int) — Cumulated number of downloads of the model since its creation.

[John6666]

Also, let’s specify downloads_all_time with the expand=[“createdAt”, “likes”, “downloads”, “downloadsAllTime”] argument. Otherwise, it will usually return None.

HfApi Client

expand (List[ExpandModelProperty_T], optional) — List properties to return in the response. When used, only the properties in the list will be returned. This parameter cannot be used if full, cardData or fetch_config are passed. Possible values are "author", "baseModels", "cardData", "childrenModelCount", "config", "createdAt", "disabled", "downloads", "downloadsAllTime", "gated", "gguf", "inference", "inferenceProviderMapping", "lastModified", "library_name", "likes", "mask_token", "model-index", "pipeline_tag", "private", "safetensors", "sha", "siblings", "spaces", "tags", "transformersInfo", "trendingScore", "widgetData", "usedStorage" and "resourceGroup".

[Ph94]

Thanks that seemed to solve the issue.