Coinbase fine-tunes Qwen3.5-9B for fraud prevention, beating frontier models at lower cost
garrytan · x · 2026-10-13
- Coinbase CEO Brian Armstrong says the company fine-tuned open-weight Qwen3.5-9B on proprietary internal data, outperforming frontier models on a specialized fraud-prevention task at much lower cost and higher speed.
- Coinbase exposes a "/fine-tune-and-serve-llm" skill in its internal harness so employees can spin up custom models easily.
- River AI, founded by ex-DeepMind/OpenAI/xAI researchers, argues custom models beat top frontier models on internal workflows at 1% of the cost, deploys its API on-prem on customer clusters, and predicts 99%+ of future LLM tokens will come from efficient custom models.
More from Models
- Gemini 4 Argon pricing ($2/$10) spotted on Google's SKU page, global release imminent — gaganghotra_ · 2026-10-13
- A style guide for making Claude-like models write normally, and its linguistic traps — gleech · 2026-10-13
- OpenAI denies Pro 200 grandfathering to a subscriber billed through the eligibility window — IrrelevantNecessity · 2026-10-13
- autotrust's JEV-35B-NVFP4, a Qwen3.5 MoE-based decision model, hits HF trending — autotrust · 2026-10-13
- Theo benchmarks Opus 5.5, GPT-6.1 Sol and Muse Wonder Full on a bug fix — shadcn · 2026-10-13
- Claude Code at $100/month plus $20 ChatGPT: AI subscription costs stack up — dumpshoot · 2026-10-13