When was U2 released?

U2 was released on June 5, 2026 by Unisound. This is the official U2 release date tracked on LLM Stats.

How much does U2 cost?

U2 costs $0.15 per million input tokens, $0.00 per million cached input tokens, $0.30 per million output tokens through the LLM Stats API, which works with any OpenAI-compatible SDK. Across tracked providers, the lowest price is $0.15 per million input tokens via Unisound. That provider charges $0.00 per million cached input tokens.

Is U2 available via API?

Yes. U2 is available through the LLM Stats API and works with any OpenAI-compatible SDK — point your client at the gateway base URL and pass the model name. It is served by 1 provider tracked on LLM Stats.

U2 has 266 billion parameters. It was trained on 15.0 trillion tokens.

What is the license for U2?

U2 is released under the Proprietary license.

What is the knowledge cutoff date for U2?

U2 has a knowledge cutoff of January 2026, meaning it was trained on data up to that point and may not know about events after it.

U2 is available through 1 provider including Unisound.

U2 API Pricing, Context Window & Benchmarks

Name: U2
Author: Unisound

U2: API Pricing, Context Window & Benchmarks

U2 is a language model from Unisound, released in June 2026, with pricing from $0.150/M input, $0.003/M cached input, $0.300/M output.

Unisound has launched its next-generation flagship model, Unisound U2, built around three core advances: high intelligence density, high token efficiency, and a native agent architecture. Through refined data curation and dense semantic

Input

TextImage

Output

Text

U2 pricing

Providers

U2 starts at $0.150 per million input tokens and $0.300 per million output tokens via Unisound. Reused prompt prefixes cost $0.0030 per million cached input tokens.

Provider	Input $/M	Cached input $/M	Output $/M	Workload 1M + 100K	Context in / out	TTFT p50 / p95 s	Output avg / p5 c/s	Success 7d	Modalities in / out
Unisound	$0.150	$0.0030	$0.300	$0.180	—/—	0.00/0.00	—/—	0.00%(27)	/

Workload cost uses 1M input tokens plus 100K output tokens at published list prices without assuming a cache hit. Cached input is the discounted price for prompt tokens served from a provider cache. TTFT is time to first token. Output is characters per second; p5 is the sustained floor exceeded by 95% of observed requests. Success is calculated from completed versus failed requests over the trailing seven days.

U2 API

POST/v1/chat/completions

Modelu2

API key●

Prompt●

Stream

Run a request to see the response

Use it in your code

Billed at $0.15 input / $0.00 cached input / $0.30 output per 1M tokens through the LLM Stats gateway.

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://gateway.llm-stats.com/v1"
)

response = client.chat.completions.create(
    model="u2",
    messages=[
        {"role": "user", "content": "What is machine learning?"}
    ]
)

print(response.choices[0].message.content)

Need an API key? Create one above in the playground, or read the API documentation.

U2 latency

U2 time to first token, sustained output throughput, and failed-request rate from live API traffic over the trailing 7 days.

Provider operational metrics

Time to first token, output throughput, and failed-request rate from live API traffic

Loading chart...

U2 examples

Recent arena outputs from U2, picked from the highest-ranked matchups.

U2 license

U2 is a proprietary model available under its provider's product and API terms, has 266.0B parameters, has a knowledge cutoff of January 2026.

License: Proprietary; Hosted access
Parameters: 266.0B
Knowledge cutoff: January 2026

Proprietary license - usage restrictions apply

U2 resources

Official sources for U2: api documentation.

API documentationmaas-api.hivoice.cn