6/3 Weekly Update: Qwen3.5-9B + Auto Top-Up

Weekly Product Update: Qwen3.5-9B + Auto Top-Up

After another week of processing billions of tokens on Doubleword, here’s what we’ve shipped.

New model: Qwen3.5-9B (exclusive)

This week we’ve added Qwen3.5-9B to Doubleword — currently not available on any other providers. Qwen3.5-9B is a compact 9B parameter model designed for strong reasoning performance while remaining extremely cost-efficient, making it a strong fit for high-volume batch workloads where you still need high-quality outputs. We’re excited to be exclusively offering this new high-intelligence model to Doubleword users.

Pricing
High Priority: $0.04 / $0.35
Standard Priority: $0.03 / $0.29

Comparable performance at a fraction of the cost

Despite its size, Qwen3.5-9B punches well above its weight. It outperforms much larger models like GPT-OSS-120B and GPT-5-nano on several key tasks, including multilingual knowledge and graduate-level reasoning. For batch workloads, this means you can maintain quality while dramatically reducing cost per token.

New feature: Auto Top-Up

We’ve also shipped Auto Top-Up, based directly on customer feedback. If you’re running long or large-scale workloads, running out of credits mid-batch is painful — Auto Top-Up removes that risk. You can set a balance threshold and define a top-up amount, and Doubleword will automatically add credits when your balance drops below that level, keeping your pipelines running smoothly without manual intervention.

Try it out

If you’re already running workloads on Doubleword, now’s a great time to test Qwen3.5-9B against your current setup and let us know what you find.