Neural Magic

    Neural Magic

    Neural Magic is a software and services provider that optimizes and serves open-source large language models for production, focusing on inference efficiency (latency and throughput) and hardware utilization via model compression and a vLLM-based serving stack.

    Est 2017|Valuation $140M|Raised $50M|4 Ref Customers
    LinkedIn 19K|Website|FTEs 53|HQ Somerville, Massachusetts, US