Microsoft-Decision-1 · House #35,944

What it is

Calls itself “decision-scoring model”.

• Calls itself: decision-scoring model

“…decision-scoring model”

Where it puts itself

Claims “highest accuracy”. Publishes a comparison page.

• Superlative claim: highest accuracy

“Unlike LLMs, which are designed to generate text or reason through complex problems, decision models are purpose-built to deliver structured outputs that software can immediately act on.”

“Microsoft-Decision-1 achieved the highest accuracy in our 36-benchmark comparison”

What it costs

A free tier up to a usage limit. Prices quoted: $0.042 per million input tokens.

• Free plan: free up to a limit
• Price: $0.042 per million input tokens

“$0.042 per million input tokens, with free output.”

Who it sells to

Sells to developers.

• Sells to: developers

“By builders, for builders.”

How it runs

Hosted in the cloud. Offers an API. AI is the core of the product.

• Runs where: cloud / hosted
• API: yes
• AI role: core to the product

“Available now in Microsoft Foundry at $0.042 per million input tokens”

“…available in Microsoft Foundry and through OpenRouter”

What it works with

Names GPT-6 Sol, H2O-Lightning-4B v1.1 and 4 more on its pages.

• Names: GPT-6 Sol · H2O-Lightning-4B v1.1 · Jev · +3

“Post-trained from Qwen3.5-9B, it is about 35x faster than GPT-6 Sol at median latency, per Microsoft.”

“2.5 times quicker than H2O-Lightning-4B v1.1, the runner-up, and 35 times quicker than GPT-6 Sol.”

What it offers as proof

Live, no waitlist. Publishes a blog.

“Official Microsoft Blog -> https://blogs.microsoft.com/”