Qwen3 235B A22B Instruct 2507 API Prices Rise: Input Up 144%, Output Up 60%

What Changed

Qwen3 235B A22B Instruct 2507 has received a notable price increase. Input costs moved from $0.09 to $0.22 per 1M tokens — a +144% jump — while output pricing rose from $0.55 to $0.88 per 1M tokens, roughly a 60% increase.

Does It Matter for Your Workload?

It depends heavily on your input/output ratio. Workloads that are read-heavy (long context stuffed into prompts, retrieval-augmented pipelines, document summarization) will feel this most sharply — that input price more than doubled. Output-heavy use cases like content generation also take a hit, but the relative increase is smaller.

As a rough check: if you were pushing 10M input tokens per month, your input bill moves from $0.90 to $2.20. At 100M tokens, that's $90 vs. $220 — a meaningful line item.

What to Do Next

Before absorbing the increase, it's worth checking whether a comparable model undercuts this new pricing. The 235B parameter count with a 22B active MoE architecture sits in a specific performance tier, but several dense and MoE alternatives on our site are priced below these updated rates.

Check the Qwen3 235B A22B Instruct 2507 — live specs & price history to track further moves and run side-by-side cost comparisons against alternatives.

Bottom Line

This isn't a minor rounding adjustment — a 144% input price increase warrants a quick audit of your monthly token spend. If this model is central to your stack, now is a good time to benchmark alternatives before your next billing cycle.

Ezra, Scout AI Team