General Learning Intelligence:
The new path to AGI.
Building superintelligence from first principles.
Checks and balances summarizer
A/B Testing Analytics
Market analysis tool
ETF Trading Dashboard
Boltzbit Research Lab
AGI 2.0: General Learning Intelligence marks the new era of AI.
Artificial General Intelligence is meant to elevate humanity but we believe the race towards a singular dominant AI power is a great threat to human creativity.
Our mission is to create General Learning Intelligence (GLI) that gives everyone the chance to own their unique AI that learns from private data and intellectual properties. We believe this is path towards AGI 2.0.
Research
View allA Gradient Based Strategy for Hamiltonian Monte Carlo Hyperparameter Optimization
This paper presents a gradient-based framework for hyperparameter optimisation in Hamiltonian Monte Carlo, directly maximising convergence speed by optimising the expected log-target under the chain’s final state.
A Gradient Based Strategy for Hamiltonian Monte Carlo Hyperparameter Optimization
This paper presents a gradient-based framework for hyperparameter optimisation in Hamiltonian Monte Carlo, directly maximising convergence speed by optimising the expected log-target under the chain’s final state.
Quasi-Newton Methods for Markov Chain Monte Carlo
A Novel MCMC sampler makes use of quasi-Newton approximations from the optimization literature, that approximate the Hessian of the target distribution from previous samples and gradients generated by the sampler.
Continuous Relaxations for Discrete Hamiltonian Monte Carlo
This article discusses the recent advancements in transformation of Boltzmann machines for gradient-based Hamiltonian MonteCarlo inference.
Latest News
View allUnder the hood: The Glimpse–Boltzbit partnership
Boltzbit and Glimpse recently announced a multi-phase AI integration. This conversation provides an overview of the collaboration, the integration approach, and its practical implications.
A-TEAM INSIGHT: Glimpse Markets Partners with Boltzbit to Embed Live-learning AI into Fixed Income Workflows
The multi-phase integration focuses on augmenting human decision-making through real-time, explainable insight generation and continuous model learning, while maintaining strict data governance and buy-side control.
Glimpse Markets partners with Boltzbit on multi-phase AI integration
Partnership will bring custom data visualisation agents, insights generation and predictive analytics to Glimpse’s bond data sharing platform.
The Boltzbit end-of-year reading list 2025
With the year drawing to a close, we asked the team to share one book they’d recommend heading into 2026. The result is a reading list that’s thoughtful, diverse, and unmistakably Boltzbit. Here’s what made the cut.
THE TRADE: What AI in in financial services will look like in 2026
In this opinion piece for The TRADE, Dr. Yichuan Zhang examines how the rise of AI is reshaping financial markets and looks ahead to 2026 and beyond.
Boltzbit appoints Leonid Belov as Head of Client Success
The firm expands leadership team to support growing demand for real-time, adaptive AI across financial services.
What customers say
"We’ve always been big proponents of interoperability and integrated solutions to bring true efficiency to bond primary markets. Our partnership with Boltzbit is a reflection of that strategy. Rather than rely on traditional Management Information reports, Boltzbit delivered a solution which allows us to interrogate the data via natural language queries, supported by dynamic reports."
Mark Russell
Head of Fixed Income Strategy,
TP ICAP
“We founded New Issue IQ to make the primary markets more efficient, so investors can capture opportunities they'd otherwise miss. Boltzbit's technology lets us structure deal data the moment it's announced, turning primary market chaos into something clients can actually use. This capability enhances the experience for our corporate bond clients today and positions us to transform new issue workflows across loans and other asset classes that still rely on outdated systems.”
John Tokarowski
Founder & CEO
New Issue IQ
"Quod’s vision is to provide an architecture where intelligence is modular, interoperable, and easy to adopt. Clients want flexibility, be it the ability to use Quod’s native AI tools or integrate complementary technologies like Boltzbit."
Medan Gabbay
Co-CEO
QUOD Financial
Features
View all
Chat
Space
LLMs
API
