---
format: "aidr-story-markdown/v1"
id: "de25b1e6ffe8cbdd6940cf9767429da383cc71c4f5ba21d2d14bd39acf609d1f"
canonical_url: "https://aidr.today/de25b1e6?lang=en"
title: "Coinbase Beats Frontier Models in Fraud Detection Using Qwen3.5-9B"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-10-12T00:37:30.000Z"
category: "Industry"
topics: ["llm"]
source_urls: ["https://huggingnews.com/ai/coinbase-beats-frontier-models-in-fraud-detection-using-qwen35-9b-3179a4fd","https://x.com/brian_armstrong/status/2109412046244913643","https://marketbrief.now/ai/coinbase-beats-frontier-models-in-fraud-detection-using-qwen35-9b-3179a4fd"]
summary: "Coinbase fine-tuned the open weight Qwen3.5-9B model on proprietary data to identify fraudulent activity, outperforming generalist frontier models in speed and cost. The company has implemented an internal harness allowing employees to train their own models for tasks such as payments risk review."
---

# Coinbase Beats Frontier Models in Fraud Detection Using Qwen3\.5\-9B

> [Open the canonical story](<https://aidr.today/de25b1e6?lang=en>)

**Published:** 2026-10-12T00:37:30.000Z
**Category:** Industry
**Topics:** llm

## Summary

Coinbase fine\-tuned the open weight Qwen3\.5\-9B model on proprietary data to identify fraudulent activity, outperforming generalist frontier models in speed and cost\. The company has implemented an internal harness allowing employees to train their own models for tasks such as payments risk review\.

## Sources

- [Story source](<https://huggingnews.com/ai/coinbase-beats-frontier-models-in-fraud-detection-using-qwen35-9b-3179a4fd>)
- [Story source](<https://x.com/brian_armstrong/status/2109412046244913643>)
- [Story source](<https://marketbrief.now/ai/coinbase-beats-frontier-models-in-fraud-detection-using-qwen35-9b-3179a4fd>)