Press Release

ThirdAI’s BOLT Deep Learning Engine Achieves Exceptional Performance on AMD EPYC™ Processors, Surpassing State-of-the-Art GPU Models

June 13, 2023 (Houston, TX) – ThirdAI, a pioneering startup dedicated to democratizing
machine learning capabilities, has successfully completed benchmarking of its BOLT deep
learning engine on AMD EPYC™ 9004 Series processors as well as the latest 9004 Series
AMD EPYC processors based on the “Zen 4c” architecture. These benchmarking experiments
showcase BOLT’s exceptional performance acceleration on cutting-edge AMD EPYC CPUs
versus current state-of-the-art models trained on NVIDIA GPUs.

In a series of rigorous tests, ThirdAI BOLT engine’s acceleration on AMD EPYC CPU-based
machines outperformed well-established baselines across diverse machine learning tasks on

Graph node classification (a technique to detect fraud or scams within social media): using
the open-source Yelp-Chi fraud detection dataset, BOLT surpassed accuracy measurements
against well-established graph neural network baselines GCN (Graph Convolutional Networks)
and GAT (Graph Attention Networks).

Trained on Yelp-Chi fraud dataset:
BOLT: 91.1% accuracy with training time of 10s on AMD EPYC 9754 CPU
GCN: 63.6% accuracy with training time of 150s
*training time 4.51s on NVIDIA A100 GPU
GAT: 81.4% accuracy with training time of 200s
*training time 22.5s on NVIDIA A100 GPU

Sequence 2 Sequence (models trained for purposes such as language translation): evaluation
on Multi-30K Translation Dataset from English to German, BOLT far surpassed translation
accuracy against standard LSTM (Long Short-Term Memory) GPU-trained model.

Trained on Multi-30K Translation Dataset from English to German:
BOLT: 39% accuracy with inference of 10ms (AMD EPYC 9754 CPU)
LSTM Seq2Seq: 20.31% accuracy with inference 29.3ms (AMD EPYC 9654 CPU)
*LSTM Seq2Seq: 20.31% accuracy with inference of 12ms (NVIDIA A100 GPU)

Text Classification (technique for purposes such as sentiment analysis or intent prediction):
evaluating ThirdAI BOLT against state-of-the-art pre-trained GPU-based model RoBERTa,
BOLT achieved similar accuracy on two representative datasets, but at a fraction of the training
time. This represents nearly a 200x speedup over GPUs.

Trained on Yelp Polarity and Amazon Polarity:
BOLT (AMD EPYC 9654 CPU): 92.3% accuracy (not pre-trained) with 130s total training
RoBERTa (AMD EPYC 9654 CPU): 94.5% accuracy (pre-trained/tuned) with 9.1 hrs
training time
*RoBERTa (NVIDIA A100 GPU): 1.77 hrs total training time

“These benchmarking experiments underscore the remarkable performance and acceleration
delivered by ThirdAI’s BOLT engine on the latest 4 th Gen AMD EPYC processors. With the
power of ThirdAI’s dynamic sparsity and AMD EPYC high performance cores and memory, we
can deliver the fastest AI acceleration in the industry,” says Anshumali Shrivastava, CEO of ThirdAI.

“The results reinforce ThirdAI’s commitment to democratizing machine learning capabilities and
enabling cost-effective training and deployment of large language models.”

About ThirdAI:
ThirdAI is on a mission to make sophisticated large language models (LLMs) and other cutting-
edge AI technologies accessible for everyone. Our goal is to build customized, private AI that is
trained on commodity hardware with ultra-low latency inference for every organization. ThirdAI’s
innovative technology is the result of 10 years of research and development in finding
fundamental ways to make deep learning more efficient. ThirdAI does not require GPUs, TPUs,
or custom ASIC to build its AI solutions. Our technology has applications in search,
recommendations, chatbots, sentiment analysis, and more.

For more information visit
AMD, the AMD arrow logo, EPYC and combinations thereof are trademarks of Advanced Micro
Devices, Inc.

Press Release

ThirdAI and Lumenci Forge Strategic Partnership to Revolutionize Legal Processes with LLMs

June 27, 2023 (Houston, TX)ThirdAI, a trailblazing company dedicated to democratizing AI through the development of private, efficient, and affordable AI solutions and large language models (LLMs) trained on CPU hardware, is delighted to announce a strategic partnership with Lumenci, a renowned technology consulting and analytics firm specializing in the legal and Intellectual Property (IP) space. This collaboration marks a significant milestone in the advancement of IP and legal process automation, paving the way for a new standard in legal teams’ efficiency worldwide.

The intricate nature of legal processes often entails an overwhelming amount of text and documentation. As such, they serve as prime candidates for harnessing the power of LLMs and generative AI technologies. LLMs possess the remarkable ability to process vast volumes of information, resulting in faster and actionable insights that drive more favorable business outcomes. Moreover, data privacy and confidentiality are of paramount importance. To address organizations’ needs, LLMs must constantly adapt their understanding to meet the evolving needs of domain experts.

ThirdAI’s groundbreaking “dynamic sparsity” technology empowers LLMs to be built, fine-tuned, and deployed in an air-gapped, fully private environment using commodity hardware. Additionally, individual customers gain the flexibility to continually refine LLM models to align with their evolving needs and understanding. This partnership offers customers the opportunity to leverage and customize cutting-edge LLMs, enabling the automation of legal and IP processes while adhering to regulatory standards.

“ThirdAI is thrilled to join forces with Lumenci, ushering in a new era of AI accessibility and privacy in the legal ecosystem,” said Anshumali Shrivastava, Founder & CEO of ThirdAI. “Lumenci’s deep domain expertise in the legal and IP space, their customer-centric approach, and their commitment to enhancing IP and legal processes align seamlessly with our core vision. Their requirements serve as a perfect testament to how LLMs should be deployed in the legal and IP space.”

Through this strategic partnership, Lumenci’s clients will experience a streamlined transformation of their legal processes, unlocking unprecedented efficiency and value.

“We are excited to collaborate with ThirdAI, a pioneering leader in large language models for enterprise applications,” stated Harish Daiya, CEO of Lumenci. “The current LLM ecosystem presents challenges in our domain, particularly in terms of privacy and fine-tuning. ThirdAI’s simplified ecosystem on commodity infrastructure, coupled with their advanced fine-tuning and customization capabilities, was an obvious choice for us. The future is here, and we are proud to empower our attorney customers with unparalleled benefits, including efficient legal document review, drafting, patent research & high quality IP data analytics—all while maintaining the utmost privacy and confidentiality.”

About ThirdAI

ThirdAI is on a mission to democratize sophisticated large language models (LLMs) and other cutting-edge AI technologies.Their goal is to build customized, private AI models trained on commodity hardware, offering ultra-low latency inference capabilities to every organization.ThirdAI’s technology has applications in search, recommendations, chatbots, virtual assistants, neural databases, sentiment analysis, graph neural networks and more. By partnering with Lumenci, ThirdAI extends its mission to make AI accessible to the legal industry, revolutionizing the way legal processes are handled and enabling legal teams to operate more efficiently.

About Lumenci

Lumenci is a prominent full-service technology consulting firm with a strong focus on the legal and IP sectors. Leveraging their domain expertise and automation capabilities through large language models (LLMs), Lumenci generates significant value for legal and IP departments. With offices in Austin, the Bay Area, New York, and Gurugram, India, Lumenci serves a diverse client base across the United States and abroad.