vllm-mlx
OpenAI and Anthropic compatible LLM inference server for Apple Silicon, powered by MLX
vllm-mlx has been downloaded 77,398 times in total on PyPI, including 6,100 in the last 30 days. The latest version is 0.5.0, released Sep 17, 2026. It is distributed under the Apache-2.0 license.
Version0.5.0
Downloads
77.40k
LicenseApache-2.0
Authorvllm-mlx contributors
Downloads
Weekly, last 90d.
Includes CI traffic.
VersionsTotal0.*
Range
View
Granularity
Group by
CI traffic
Stack: OffCI: Included2 / 14 series
Selected total61.42k
77.4kAll-time
6.1kLast 30 days
99Last 24 h
<0.01/sPer second
Sponsored
Sponsorships keep pepy free to read
Version distribution
Share of downloads by released version. Computed over the last quarter.
- 0156.6%
0.4.1
17.4k downloadsDownloads17.4k56.6% - 0217.1%
0.5.0
5.2k downloadsDownloads5.2k17.1% - 0310.6%
0.2.5
3.3k downloadsDownloads3.3k10.6% - 049.3%
0.4.0
2.8k downloadsDownloads2.8k9.3% - 052.4%
0.3.0
742 downloadsDownloads7422.4% - 060.7%
0.2.6
226 downloadsDownloads2260.7% - 070.6%
0.2.8
184 downloadsDownloads1840.6% - 080.6%
0.2.9
169 downloadsDownloads1690.6% - 092.1%
Other
645 downloadsDownloads6452.1%
Guess the next day
Thirteen recent days of vllm-mlx downloads. Drag the green handle on the right to guess where day fourteen lands.