SubQ Savant

The first long-context language model for data-intensive work.

SubQ is the first model built for multi-million token reasoning, allowing enterprises to work across full repositories, financial filings, and contract archives for a fraction of the cost.

Built by researchers and engineers from
The problem

Today’s LLMs are not scalable. Businesses are forced to choose between quality, cost and security while building complex systems to work around the small “effective” context windows of today’s models.

Use cases

Data-heavy workloads become possible.

Whole Artifact Reasoning

Analyze entire datasets at once: an entire GitHub repo, all legal documentation, every vendor contract, or years of SEC filings, without chunking, compression, or context loss.

Long Horizon Agents

Agents that hold the full task in context and remember long sequences of events, collapsing brittle multi-agent orchestration into one coherent run.

Search & Retrieval

Surface material insights that you can trust from across an entire company's IP, code, or documents without risk of hallucination.

All your context.
Always available.

Reason across millions of tokens in one prompt: entire repos, whole artifacts, and long-running agent state, with context to spare.

  • Five enterprise vendor contracts

    Each a 500-page master agreement

  • Three years of SEC filings

    A Fortune 500 10-Ks and 10-Qs

  • Python source code

    The entire 3.13 standard library

~ Approximate token counts.

The breakthrough

The most compute-efficient model.

Today's transformer-based LLMs waste compute by processing every possible relationship between tokens, but only a small fraction of these relationships matter.

SubQ is built differently. Using a proprietary algorithm called Subquadratic Sparse Attention (SSA), it isolates the tokens and relationships that matter, ensuring compute is used efficiently.

At 2M tokens, SubQ uses 128x less compute than frontier models.

0 128K 512K 1M 2M TOKENS 8× 31.5× 64.5× 128× 1,008 PFLOP 7.8 PFLOP Frontier Models (Dense Attention) SubQ (SSA)
Early access

Request API Access.

By submitting this form, you agree to our privacy policy and consent to receive marketing communications from Subquadratic. You can unsubscribe at any time.

Building a radically more intelligent world

Subquadratic is a frontier AI research company that believes the architecture layer is the highest-leverage surface in AI. While other major labs focus on incremental improvements to Transformer models, we're pushing foundational change at the architecture level to build models and products that scale efficiently.

Built by researchers and engineers from
  • Meta
  • Google
  • Oxford
  • Cambridge
  • MIT
  • BYU
  • DoorDash
  • Amazon
  • LlamaIndex
  • TwelveLabs
  • Airbnb