gpudiff the public record of change in GPU cloud pricing
Follow the diffs: RSS · Free API · Sponsor this site

qwen3-vl-235b-a22b-thinking API price — 3 gateways compared

from $0.40 per million input tokens · 2.4× spread between gateways · updated hourly

Gateway$ in /MTok$ out /MTok ContextInput price historyModel ID
OpenRouter fees$0.40$4.00131,072qwen/qwen3-vl-235b-a22b-thinking
Glama$0.40$4.00131,072since 2026-08-20alibaba/qwen3-vl-235b-a22b-thinking
Novita$0.98$3.95131,072since 2026-08-20qwen/qwen3-vl-235b-a22b-thinking

Fees beyond the token price

Changelog for qwen3-vl-235b-a22b-thinking

qwen3-vl-235b-a22b-thinking pricing FAQ

How much does qwen3-vl-235b-a22b-thinking cost per million tokens?

The cheapest tracked gateway charges $0.40 per million input tokens and $3.95 per million output tokens (Glama). Prices refresh hourly.

Which gateway is cheapest for qwen3-vl-235b-a22b-thinking?

Glama for input tokens and Novita for output tokens, among the 3 gateways we track.

Do qwen3-vl-235b-a22b-thinking prices differ between gateways?

Yes — the most expensive tracked gateway charges 2.4× the cheapest for input tokens. Identical model, identical tokens, different bill.

← All LLM prices · API