Qwen3.6 27B API Prices Cut 50%: Input Drops to $0.3, Output to $2.00 per 1M Tokens

What Changed

Qwen slashed pricing on Qwen3.6 27B across both token directions. Input cost dropped from $0.6 to $0.3 per 1M tokens — a clean -50% cut. Output pricing moved from $3.60 to $2.00 per 1M tokens, a reduction of roughly 44%.

Does It Matter for Your Workload?

It depends heavily on your input/output ratio. For retrieval-augmented generation (RAG) pipelines and classification tasks — where you're pumping in long contexts but getting short responses — the input cut is the headline win. A workload burning 10M input tokens per month just dropped from $6.00 to $3.00. That's real money at scale.

Output-heavy use cases (long-form generation, code completion, multi-turn chat) benefit less proportionally. The output price moves from $3.60 to $2.00 per 1M tokens, which is still a meaningful saving but not the symmetrical halving the input side delivers.

Context

This cut follows a broader industry pattern of aggressive mid-tier model pricing compression. At $0.3 input / $2.00 output, Qwen3.6 27B sits in a competitive band for a model of this size. Whether it beats your current stack depends on quality-per-dollar for your specific tasks — benchmark comparisons matter here, not sticker price alone.

Next Steps

If you're evaluating whether to switch or scale usage, check the Qwen3.6 27B — live specs & price history page for up-to-date benchmark scores and a full pricing timeline to compare against alternatives in the same tier.

Bottom line: the -50% input cut is significant enough to warrant re-running your cost model. The output reduction is a bonus, not the main event.

Ezra, Scout AI Team