LLM Reference

Unisound U2 on Unisound Token Hub

Unisound U2 · Unisound

Serverless

Last refreshed 2026-06-20. Next refresh: weekly.

Why use Unisound U2 on Unisound Token Hub?

Unisound Token Hub offers Unisound U2 with competitive pricing. Unisound Token Hub is Unisound's MaaS platform for text, speech, and vision models.

Input / 1M
-
Output / 1M
-
Cache
Not sourced
Batch
Not sourced

Setup recipe

Docs fallback
Install
Use the provider REST API or SDK
Auth
Create a provider API key
Call
model: u2
Model ID
u2

Request example

Curated snippets for this provider are not sourced yet. Use Unisound Token Hub documentation with model ID u2.

Gotchas

  • Use provider model ID "u2", not the LLMReference slug "unisound-u2".

Capabilities

ReasoningFunction CallingTool Use

About Unisound U2

Unisound U2 is a native agentic large language model announced June 8, 2026, by Chinese AI company Unisound (云知声). Designed for multi-step autonomous task execution, it uses native reasoning-path distillation for task decomposition, execution, verification, and optimization. The official Token Hub page lists a sparse MoE architecture with 266B total parameters and 10B active parameters, plus API access through Unisound Token Hub. Emphasis is on coding, tool orchestration, and long-horizon agent workflows. Unisound self-reports GPQA Diamond 87.9, SWE-bench Verified 75.0, and Claw-Eval 76.9; treat those benchmarks as vendor-reported until independently verified. Context window is not disclosed.

FAQ

What API model ID do I use for Unisound U2 on Unisound Token Hub?

Use the model ID u2 when calling Unisound Token Hub's API.

Who created Unisound U2?

Unisound U2 was created by Unisound as part of the Unisound U2 model family.

Is Unisound U2 open source?

Unisound U2 is not open source; the seed data lists it as proprietary.

Get Started

Model Specs

Released2026-06-08
Parameters266B
ArchitectureMixture of Experts