SubQ Snap

Released
2026-10-12
Last refreshed
2026-10-13
Status
Researched today
ProprietaryCommercial use: conditionalMultimodalRAGAgentsLong contextVisionJSON / Tool useAgentsVision
Subquadratic releases · 2 in the last 12 months · this family litChangelog →
Specifications
Family
SubQ Snap
Released
2026-10-12
Context
384K
Openness
Proprietary
License
ProprietaryCommercial use: conditional
Weights
Not released
Code
Unknown
Created by

Efficiency is Intelligence

Founded 2026
Website
Pricing
Output / 1M
Free
Input / 1M
$0.050

Cheapest of 1 route · SubQ API

About

SubQ Snap is Subquadratic's long-context multimodal decision model, built on Subquadratic Sparse Attention (SSA) with a decision head that maps the input to a probability distribution over the allowed answers. It accepts text and up to 4 images per request within a 384K-token (393,216) context window and answers 1–16 named noul (yes/no), choice, or score questions per call. Released 2026-10-12 in public preview on the SubQ Decisions API as subq-snap-preview. Subquadratic reports 61.7 on Decision Index 0.2.1 (vs 57.9 for Jev) and 82.9% on Bespoke Nimble 13; these are vendor-reported. Not a chat model.

Providers(1)

Compare API pricing across 1 tracked provider route for input and output tokens, batch, and cached reads when available.

ProviderInput / 1MOutput / 1MRoute
SubQ API$0.050Free
Serverless

Capabilities

VisionMultimodalStructured Outputs

Benchmark peer barsfor RAG

No task-mapped benchmark peers are available for this model yet.

Migration checks

No linked migration route is available for this model yet.

API versions

subq-snap-previewSubQ Snap Preview