Union Alpha Model

A free multimodal model built for research, coding, and agentic workflows — with a 262K context window, text + image input, and frontier-level performance.

Free 262K Context Text + Image → Text OpenAI Compatible Released Sep 16, 2026

Specifications

Model IDstealth/union-alpha
ProviderStealth (third-party, operated independently of OpenRouter)
ReleasedSeptember 16, 2026
PriceFree
Context window262,144 tokens
Max output131,072 tokens
Input → OutputText + Image → Text
Tool callingSupported (tools / tool_choice)
Structured outputSupported (response_format), JSON schema not enforced
P50 latency~17.6 seconds
P50 throughput~14 tokens / second
3-day uptime98.08%
Data policyPrompts and completions may be retained by the provider but are not used for training

Quick Start

Union Alpha uses the OpenAI-compatible API via OpenRouter. Use the model ID stealth/union-alpha in any client.

cURL

curl https://openrouter.ai/api/v1/chat/completions \
  -H "Authorization: Bearer $OPENROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "stealth/union-alpha",
    "messages": [
      {"role": "user", "content": "Explain the transformer architecture in simple terms."}
    ]
  }'

Python

import openai

client = openai.OpenAI(
    base_url="https://openrouter.ai/api/v1",
    api_key="YOUR_OPENROUTER_API_KEY",
)

response = client.chat.completions.create(
    model="stealth/union-alpha",
    messages=[{"role": "user", "content": "Explain the transformer architecture."}],
)
print(response.choices[0].message.content)

Cline / OpenCode / Hermes Agent: just set the model ID to stealth/union-alpha and point the API base to https://openrouter.ai/api/v1.

Full setup guide →

Complete API Example

Real request and response from Union Alpha via Tokenra.io API proxy.

Request

POST https://tokenra.io/v1/chat/completions
Content-Type: application/json

{
  "model": "union-alpha",
  "messages": [
    {
      "role": "user",
      "content": "what model"
    }
  ],
  "max_tokens": 1024,
  "stream": false
}

Response

{
  "id": "gen-1789580899-b6F0U6o1wxcGXYak2Ahb",
  "object": "chat.completion",
  "created": 1789580899,
  "model": "stealth/union-alpha",
  "provider": "Stealth",
  "choices": [
    {
      "index": 0,
      "finish_reason": "stop",
      "message": {
        "role": "assistant",
        "content": "I am Union Alpha. My maker is currently anonymous."
      }
    }
  ],
  "usage": {
    "prompt_tokens": 1,
    "completion_tokens": 12,
    "total_tokens": 13,
    "cost": 0
  }
}

Key points: Zero cost, OpenAI-compatible format, and clean responses. Use union-alpha as model ID on Tokenra, or stealth/union-alpha on OpenRouter.

Who's Using Union Alpha

Token volume on OpenRouter since launch. Real production usage — not benchmarks.

omp
965M
Cline
342M
Hermes Agent
317M
ZCode
206M
DeepSeek Harness
190M

What to Watch Out For

Free doesn't mean perfect. Here's what you should know before relying on Union Alpha in production.

Frequently Asked Questions

Is Union Alpha free?

Yes. Union Alpha is currently free to use on OpenRouter with no per-token cost. It's a 262K-context multimodal model available at stealth/union-alpha.

What is Union Alpha?

Union Alpha is a free multimodal large language model on OpenRouter (stealth/union-alpha), built for research, coding, and agentic workflows. It supports text and image inputs with a 262K token context window.

What is the context length?

262,144 tokens. Maximum output is 131,072 tokens.

Does Union Alpha support tool calling?

Yes. It supports OpenAI-compatible function calling (tools / tool_choice) and structured outputs via response_format. JSON schema is supported but not enforced.

Who made Union Alpha?

Union Alpha is developed and operated by an anonymous third-party provider labeled "Stealth" on OpenRouter. No company name or research paper has been disclosed.

How is Union Alpha different from Ox Alpha?

Both are Stealth models on OpenRouter, but they differ in capabilities, pricing, and performance. See our detailed comparison for benchmark data.

Why is Union Alpha slow?

At ~17.6s P50 latency and ~14 tok/s throughput, Union Alpha prioritizes depth over speed. It is best suited for long-form generation and batch workloads rather than interactive chat.

All Guides