PrismML is a Pasadena, CA edge-AI company spun out of Caltech that builds extremely compressed large language models which run on phones, laptops, and embedded devices without sacrificing performance. It emerged from stealth on March 31, 2026 with $16.25M in SAFE and seed funding from Khosla Ventures, Cerberus Ventures, and Caltech, plus compute grants from Google and Caltech, and is built on proprietary Caltech intellectual property. Its flagship 1-bit Bonsai 8B model compresses memory from ~16GB to ~1.15GB, boosts inference speed roughly 8x, and cuts energy use by up to 80% while staying close to 16-bit performance (running ~44 tokens/s on an iPhone). In April 2026 it introduced the Ternary Bonsai family — 8B, 4B, and 1.7B models using a 1.58-bit ternary weight representation (-1, 0, +1) for a ~9x memory reduction. The founding team is led by CEO Babak Hassibi (Caltech professor), with Sahin Lale and Omead Pooladzandi as co-heads of research and Reza Sadri as VP of Strategy.

Company Profile

Founders & Key People

Investors

Products

• 1-bit Bonsai 8B (world's first commercially viable 1-bit LLM)
• Ternary Bonsai model family (1.58-bit, 8B / 4B / 1.7B parameters)

Competitors

AIEdge AI1-bit LLMModel CompressionCaltech SpinoutStealth Exit 2026

Private-company numbers are not real-time. Reflects publicly disclosed valuations from press releases, news reports, and tender offers as of recent date. Refreshed quarterly.