---
title: "Microsoft's $0.042 decision model runs on Alibaba's free Qwen, and Redmond plans to replace it"
description: "Microsoft-Decision-1 launched October 9 at $0.042 per million input tokens on Alibaba's Qwen3.5-9B weights that cost Redmond nothing. The company behind the base is named in a US intelligence advisory, and Microsoft has already said it will swap the base out."
publisher: "The Inference"
section: "Frontier"
published: 2026-10-11T17:27:50.333Z
modified: 2026-10-11T17:27:50.333Z
canonical: https://theinference.org/article/microsoft-s-0-042-decision-model-runs-on-alibaba-s-free-qwen-and-redmond-plans-to-replace-it
language: en
keywords: "Microsoft, Alibaba, Open Weights, Qwen, Model Pricing"
---

# Microsoft's $0.042 decision model runs on Alibaba's free Qwen, and Redmond plans to replace it

> Microsoft-Decision-1 launched October 9 at $0.042 per million input tokens on Alibaba's Qwen3.5-9B weights that cost Redmond nothing. The company behind the base is named in a US intelligence advisory, and Microsoft has already said it will swap the base out.

## Alibaba's free weights, wearing a Redmond badge

[Microsoft](https://theinference.org/markets/companies/microsoft)'s cheapest new AI model is a Chinese lab's open weights wearing a Redmond badge. Microsoft-Decision-1 [went into general availability on October 9](https://www.marktechpost.com/2026/10/09/microsoft-ai-releases-microsoft-decision-1-a-qwen3-5-9b-decision-scoring-model/): a model that returns a probability for each fixed option instead of writing text, built for routing customer inquiries and grading what AI agents propose to do [1]. Underneath sits [Alibaba](https://theinference.org/markets/companies/alibaba)'s Qwen3.5-9B, a 9-billion-parameter open-weight model, retrained on public and synthetic data to score options in a single inference pass [1][2]. More than 100 decision models now compete for attention after Jev's arrival three weeks ago, and two of the most prominent, Microsoft's and H2O.ai's, are Qwen underneath [3].

## The price undercuts the American incumbent

Decision-1 costs $0.042 per million input tokens, output free, more than 20x cheaper than [OpenAI](https://theinference.org/markets/companies/openai)'s GPT-6 Sol in text classification and less than half the input price of OpenAI's Luna decisions endpoint [1][2][3]. Microsoft claims the best accuracy in its own [36-benchmark comparison](https://xenospectrum.com/en/microsoft-decision-one-benchmarks-calibration/), 83.5% across 147,137 questions, and a calibration score of 92.2 that trails Jev's 93.7 [1][2]. The headline speed claim, roughly 35x GPT-6 Sol, is measured against a general-purpose model; against GPT-6 Luna Decisions, a decision model, the same published figures give about 3.5x [1].

*Chart: **Microsoft's published 85 ms met a third party's 459 ms measurement** Horizontal bars comparing median response times: Decision-1 measured by Microsoft at 85 ms and by JevBench at 459 ms, Jev at 240 ms, Quyet-1.0-Large at 380 ms, GPT-6 Sol at 3010 ms.*

*Median response time in milliseconds. Microsoft's 85 ms was measured in-region on Foundry; its competitor figures use JevBench v1.6.1 adjusted medians from its 36-benchmark comparison. JevBench's 459 ms for Decision-1 comes from its own 124-request test through Azure Foundry on October 10, 2026, with different questions and network paths. Source: Xenospectrum [1]. [1]*

## Redmond pre-announced the expiry date

Microsoft says it will "soon" rebase Decision-1 on its own MAI models and OpenAI's, for reasons not disclosed [3][2]. The base that makes the price possible costs Redmond nothing and now carries a stated end date. What Alibaba holds instead is ubiquity, and a security file. On September 9 [the NSA, FBI and CISA jointly named Alibaba among six Chinese firms conducting industrial-scale distillation of US frontier models](https://thenextweb.com/news/nsa-fbi-cisa-advisory-chinese-ai-distillation) since late 2024, accusing it of distilling Claude and GPT-5 in late 2025 to improve the Qwen family [4][5]. [Anthropic](https://theinference.org/markets/companies/anthropic)'s [September report attributes more than 151 million distillation activities to Alibaba](https://interestingengineering.com/ai-robotics/anthropic-accuses-chinese-ai-firms) between May and July 2026; the six companies did not respond to requests for comment, and the allegations remain contested [6].

*Chart: **Alibaba's capex tripled to $18.2B while operating income fell 62.5% in FY2026** Columns over six fiscal years showing Alibaba's capex rising from $6.3B in FY2021 to $18.2B in FY2026 while operating income falls from $19.4B in FY2025 to $7.3B in FY2026.*

*Company-wide figures in USD billions, fiscal years ending March 31, from Alibaba's SEC filings, retrieved October 11, 2026 [7]. FY2026 operating income reflects the AI spending surge; not a cloud-segment figure. [7]*

The spending behind Qwen shows up in the accounts: [Alibaba's capex reached $18.2B in the fiscal year ended March 2026](https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0001577552), up 58%, while company-wide operating income fell 62.5% to $7.3B [7]. Chairman Joe Tsai puts the AI build at about $25 billion of e-commerce free cash flow a year, with infrastructure capex doubling three years running [8].

## Ubiquity is Alibaba's best argument

Free weights seed the ecosystem that buys the cloud, and the demand is visible. [Ecosia dropped Mistral this week over quality problems and overloaded servers](https://cybernews.com/ai-news/ecosia-drops-mistral/) and is now weighing Qwen, GLM and Kimi for its search AI [9]. [Tsai pitched open source to a Turin audience last week as Europe's only path to AI sovereignty](https://theinference.org/article/tsai-sells-europe-open-weights-alibaba-s-safety-record-is-seven-results-in-238-releases) [8]. But open weights cut both ways, as Tsai himself explained: anyone can take the recipe, retrain it and stop depending on the original developer, which is exactly what Microsoft has pre-announced [8].

## What decides it

The open question is whether Qwen's ubiquity converts into Alibaba Cloud revenue before Microsoft's rebase or a US policy action ends the free run. Alibaba's next quarterly results, and whatever date Microsoft sets for the rebase, will tell. [Free gets adopted. It does not get paid.](https://theinference.org/article/china-is-closing-the-ai-gap-profits-are-the-harder-test)

## Takeaway

Microsoft's cheapest decision model rides Alibaba's free Qwen weights, a base now named in a US intelligence advisory, and Redmond has pre-announced its replacement.

## Sources

1. [Xenospectrum, Microsoft's Decision-1 AI Scores Options Instead of Writing Text, With a Claimed 35x Speedup Over GPT-6 Sol, 10 October 2026](https://xenospectrum.com/en/microsoft-decision-one-benchmarks-calibration/)
2. [MarkTechPost, Microsoft AI Releases Microsoft-Decision-1: A Qwen3.5-9B Decision-Scoring Model, 9 October 2026](https://www.marktechpost.com/2026/10/09/microsoft-ai-releases-microsoft-decision-1-a-qwen3-5-9b-decision-scoring-model/)
3. [The Register, Microsoft leans on open weight model from Chinese AI lab to challenge Jev, 10 October 2026](https://www.theregister.com/ai-and-ml/2026/10/10/microsoft-leans-on-open-weight-model-from-chinese-ai-lab-to-challenge-jev/5302473)
4. [The Next Web, US intelligence advisory names six Chinese AI firms and lists the US models each one targeted, 9 September 2026](https://thenextweb.com/news/nsa-fbi-cisa-advisory-chinese-ai-distillation)
5. [BleepingComputer, US says Chinese firms extracted billions of tokens from frontier AI models, 9 September 2026](https://www.bleepingcomputer.com/news/security/us-says-chinese-firms-extracted-billions-of-tokens-from-frontier-ai-models/)
6. [Interesting Engineering, Anthropic accuses Chinese AI firms of secretly using Claude to train their AI models, 10 October 2026](https://interestingengineering.com/ai-robotics/anthropic-accuses-chinese-ai-firms)
7. [Alibaba, annual figures from its SEC filings, retrieved 11 October 2026](https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0001577552)
8. [The Next Web, Alibaba's Joe Tsai says open source is Europe's only way to AI independence, 7 October 2026](https://thenextweb.com/news/alibaba-joe-tsai-open-source-ai-sovereignty-europe-wave-by-vento)
9. [Cybernews, European search engine Ecosia drops Mistral and eyes Chinese AI models, 10 October 2026](https://cybernews.com/ai-news/ecosia-drops-mistral/)


---

Microsoft's $0.042 decision model runs on Alibaba's free Qwen, and Redmond plans to replace it — The Inference. Canonical: https://theinference.org/article/microsoft-s-0-042-decision-model-runs-on-alibaba-s-free-qwen-and-redmond-plans-to-replace-it