LLM ReferenceLLM Reference

SubQ 1M-Preview

subq-1m-preview

Proprietary

About

SubQ 1M-Preview is Subquadratic's first large language model, built on a fully sub-quadratic sparse-attention architecture that scales compute linearly with context length (O(n) vs. traditional O(n²)). Supports a production context window of 1M tokens (architecture tested to 12M). Achieves 81.8% on SWE-Bench Verified, 95.0% on RULER @128K, and 65.9% on MRCR v2 (8-needle, 1M). Claims 50x faster and 50x cheaper than leading frontier models at 1M context length. Available via OpenAI-compatible API with streaming and tool use support. Model is proprietary and not open-source; fine-tuning for customer-specific use cases is mentioned as a future capability.

SubQ 1M-Preview has a 1M-token context window.

Capabilities

VisionMultimodalReasoningFunction CallingTool UseStructured OutputsCode ExecutionPrompt CachingBatch APIAudioFine-tuning

Providers(1)

ProviderInput (per 1M)Output (per 1M)Type
SubQ APIServerless

Rankings

Specifications

FamilySubQ
Released2026-05-05
Context1M
ArchitectureDecoder Only
Specializationgeneral
LicenseProprietary
Trainingpretrained

Created by

Efficiency is Intelligence

Founded 2026
Website

Providers(1)