gpudiff the public record of change in GPU cloud pricing

Follow the diffs: RSS · Free API · Sponsor this site

qwen3-235b-a22b-thinking-2507 API price — 4 gateways compared

Cheapest right now: Glama at $0.23/MTok in — 1.3× cheaper than the priciest gateway.

from $0.23 per million input tokens · 1.3× spread between gateways · updated continually

Gateway	$ in /MTok	$ out /MTok	Context	Model ID
OpenRouter fees	$0.23	$2.30	131,072	qwen/qwen3-235b-a22b-thinking-2507	use
Requesty	$0.23	$2.30	262,144	deepinfra/Qwen/Qwen3-235B-A22B-Thinking-2507	use
Glama	$0.23	$2.30	131,072	alibaba/qwen3-235b-a22b-thinking-2507	use
Novita	$0.30	$3.00	131,072	qwen/qwen3-235b-a22b-thinking-2507	use

Fees beyond the token price

OpenRouter: No markup on token prices (provider pass-through), but credit purchases cost 5.5% via card ($0.80 minimum) or 5% via crypto; BYOK usage above the monthly allowance is charged 5%. source

2026-08-27 requesty raised qwen3-235b-a22b-thinking-2507-output 11.1%: 2.07 → 2.3 usd_per_mtok
2026-08-27 requesty raised qwen3-235b-a22b-thinking-2507-input 11.1%: 0.207 → 0.23 usd_per_mtok

The cheapest tracked gateway charges $0.23 per million input tokens and $2.30 per million output tokens (Glama). Prices refresh continually.

Glama for input tokens and Glama for output tokens, among the 4 gateways we track.

Yes — the most expensive tracked gateway charges 1.3× the cheapest for input tokens. Identical model, identical tokens, different bill.