Which AI provider SDKs are actually installed

· refreshed weekly

anthropic has the most-installed first-party AI API SDK at 316,942,230 downloads over the last 30 days, across PyPI and npm. Installs are not latency and not quality — they are the other question a developer asks before picking an API: is anyone actually using this.

Most-installed AI provider SDKs

#ProviderTotal PyPInpm
1anthropic316,942,230175,863,980141,078,250
2google304,892,321218,957,28985,935,032
3mistral60,708,34228,830,90531,877,437
4cohere24,965,18123,092,4701,872,711
5groq22,401,12318,836,7663,564,357
6voyage4,510,5913,608,376902,215
7xai3,766,9733,766,9730
8replicate3,414,7191,415,6811,999,038
9qwen2,604,1082,604,1080
10together2,001,6211,669,414332,207
11fireworks1,943,5571,943,54611
12cerebras1,579,9651,046,105533,860
13glm280,580280,5800
14ai21263,420263,073347
15writer247,925225,99521,930
16reka112,382112,3820
17baseten2,3852,3850
18minimax1,8381,8380

Who grew and who shrank in the last 30 days?

ProviderWasNow Change
reka70,608112,382+59.2%
ai21243,381263,420+8.2%
anthropic313,609,889316,942,230+1.1%
voyage4,516,1784,510,591-0.1%
replicate3,898,1073,414,719-12.4%
cerebras1,848,1931,579,965-14.5%
google360,201,090304,892,321-15.4%
xai4,478,8383,766,973-15.9%
openai561,278,321462,337,104-17.6%
qwen3,236,9842,604,108-19.6%
mistral76,624,75260,708,342-20.8%
baseten3,1562,385-24.4%
groq29,927,78922,401,123-25.1%
together2,695,6512,001,621-25.7%
glm388,323280,580-27.7%

Why is the openai package not in this table?

Downloads of the `openai` package are not attributed to any provider: most OpenAI-compatible providers document their own API through it, so the number measures the protocol's reach, not one vendor's usage.

What do these numbers actually measure?

Package downloads from the official PyPI and npm statistics endpoints, over the last 30 days. A download is a machine fetching the package — CI runs, Docker builds and mirrors are all in there, so this is a measure of reach, not of active users, and certainly not of quality. It is published because it is verifiable: the same two endpoints are open to anyone who wants to check the figures.

Sources: https://pypistats.org/api · https://api.npmjs.org/downloads. Latency for these same providers is measured separately — see the latency rankings.