TLDR
- GSI Technology stock jumped 155% to $12.97 on October 20 after Cornell University validated its AI chip technology
- The Gemini-I APU matched NVIDIA A6000 GPU performance while consuming 98% less energy in independent testing
- GSI’s chip processed AI tasks 5 times faster than standard CPUs with 80% reduction in processing time
- Pre-market trading showed continued gains to $16.36, though analysts maintain cautious “Sell” ratings
- The company reports Q2 earnings on October 30 with investors watching for commercial customer orders
GSI Technology stock closed at $12.97 on October 20, up 155% in a single trading session. The surge followed Cornell University’s independent validation of the company’s energy-efficient AI chip.
The stock opened the week trading around $5.08. It reached an intraday high of $17.48 before settling at the $12.97 close.
Trading volume exploded to 111 million shares compared to less than one million shares in previous sessions. Pre-market activity on October 21 pushed the stock to approximately $16.36.
Cornell researchers tested GSI’s Gemini-I Associative Processing Unit against conventional computing hardware. The study appeared at the MICRO ’25 conference with results published by ACM.
Testing showed the APU matched NVIDIA A6000 GPU throughput on retrieval-augmented generation workloads. The critical finding was energy efficiency.
Cornell Study Results
GSI’s chip delivered comparable GPU performance while using 98% less power. Researchers tested datasets from 10GB to 200GB to validate scalability.
The APU outperformed standard CPUs by processing AI tasks up to 80% faster. This represents roughly a 5x speedup on identical workloads.
CEO Lee-Lean Shu called the validation proof that compute-in-memory technology could disrupt the $100 billion AI inference market. The architecture performs calculations inside memory arrays rather than shuttling data to separate processors.
This eliminates the energy-draining bottleneck in traditional GPU and CPU designs. The approach directly addresses AI’s growing power consumption problem.
GSI has already completed its second-generation Gemini-II chip. The company claims it delivers 10 times the throughput of the first generation with lower latency.
Gemini-II targets edge AI applications including autonomous drones, satellites, and military systems where power budgets are constrained. A third chip called “Plato” is in development for embedded computing.
Analyst Caution Despite Rally
Wall Street analysts remain skeptical. Weiss Ratings maintains a “Sell” rating and MarketBeat shows analyst consensus at “Sell.”
CoinCodex flagged the stock as “very high volatility” and potentially overbought. Their models suggest possible retracement to $5, though such predictions are speculative.
GSI Technology generates approximately $22 million in annual revenue with a negative 63% net margin. The most recent quarter showed $6.3 million revenue, up 35% year-over-year, with a loss of $0.08 per share.
Current sales come primarily from niche memory chips rather than AI products. The APU technology has not yet produced large commercial revenues.
Some institutional investors increased positions. Bank of America raised its stake by 8,248% while Osaic Holdings grew holdings by 52% in recent filings.
The company maintains a current ratio of 5.79 with more cash than debt. GSI raised capital in summer 2025 to fund research and development.
The edge AI semiconductor market could reach $56 to $57 billion by 2030 according to analysts. GSI targets defense and aerospace customers and holds a U.S. Army SBIR contract.
GSI Technology reports Q2 FY2026 earnings after market close on October 30, 2025. Investors will watch for evidence of customer orders or commercial pipeline development for the APU technology.
Get 3 Free Stock Ebooks
Discover top-performing stocks in AI, Crypto, and Technology with expert analysis.
- Top 10 AI Stocks - Leading AI companies
- Top 10 Crypto Stocks - Blockchain leaders
- Top 10 Tech Stocks - Tech giants
