gpudiff the public record of change in GPU cloud pricing

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

qwen3.6-35b-a3b API price — 4 gateways compared

Cheapest right now: DeepInfra at $0.10/MTok in — 2.5× cheaper than the priciest gateway.

from $0.10 per million input tokens · 2.5× spread between gateways · updated continually

Gateway	$ in /MTok	$ out /MTok	Context	Model ID
DeepInfra	$0.10	$0.95	262,144	Qwen/Qwen3.6-35B-A3B	use
Glama	$0.14	$1.00	262,144	alibaba/qwen3.6-35b-a3b	use
OpenRouter fees	$0.15	$1.00	262,144	qwen/qwen3.6-35b-a3b	use
Novita	$0.25	$1.49	262,144	qwen/qwen3.6-35b-a3b	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

2026-09-21 openrouter listed qwen-qwen3.6-35b-a3b-input at 0.15 usd_per_mtok
2026-09-20 openrouter raised qwen-qwen3.6-35b-a3b-output 11.1%: 0.9 → 1.0 usd_per_mtok
2026-09-20 openrouter delisted qwen-qwen3.6-35b-a3b-input
2026-08-27 openrouter cut qwen-qwen3.6-35b-a3b-output 10.0%: 1.0 → 0.9 usd_per_mtok
2026-08-27 openrouter cut qwen-qwen3.6-35b-a3b-input 28.6%: 0.14 → 0.1 usd_per_mtok
2026-08-16 openrouter cut qwen-qwen3.6-35b-a3b-input 6.7%: 0.15 → 0.14 usd_per_mtok

The cheapest tracked gateway charges $0.10 per million input tokens and $0.95 per million output tokens (DeepInfra). Prices refresh continually.

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

Yes — the most expensive tracked gateway charges 2.5× the cheapest for input tokens. Identical model, identical tokens, different bill.