gpudiff the public record of change in GPU cloud pricing

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

gpt-oss-120b API price — 5 gateways compared

Cheapest right now: Glama at $0.03/MTok in — 1.7× cheaper than the priciest gateway.

from $0.03 per million input tokens · 1.7× spread between gateways · updated continually

Gateway	$ in /MTok	$ out /MTok	Context	Model ID
Glama	$0.03	$0.17	131,072	openai/gpt-oss-120b	use
Requesty	$0.03	$0.14	131,072	runware/gpt-oss-120b	use
OpenRouter fees	$0.04	$0.17	131,072	openai/gpt-oss-120b	use
DeepInfra	$0.04	$0.17	131,072	openai/gpt-oss-120b	use
Novita	$0.05	$0.25	131,072	openai/gpt-oss-120b	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

2026-09-30 openrouter listed openai-gpt-oss-120b-output at 0.17 usd_per_mtok
2026-09-30 openrouter listed openai-gpt-oss-120b-input at 0.037 usd_per_mtok
2026-09-29 openrouter delisted openai-gpt-oss-120b-output
2026-09-29 openrouter delisted openai-gpt-oss-120b-input
2026-09-19 openrouter listed openai-gpt-oss-120b-output at 0.6 usd_per_mtok
2026-09-19 openrouter listed openai-gpt-oss-120b-input at 0.15 usd_per_mtok
2026-09-18 openrouter delisted openai-gpt-oss-120b-output
2026-09-18 openrouter delisted openai-gpt-oss-120b-input
2026-09-11 novita listed gpt-oss-120b-output at 0.25 usd_per_mtok
2026-09-11 novita listed gpt-oss-120b-input at 0.05 usd_per_mtok
2026-09-10 novita delisted gpt-oss-120b-output
2026-09-10 novita delisted gpt-oss-120b-input
2026-08-31 requesty listed gpt-oss-120b-output at 0.14 usd_per_mtok
2026-08-31 requesty listed gpt-oss-120b-input at 0.032 usd_per_mtok
2026-08-30 requesty delisted gpt-oss-120b-output
2026-08-30 requesty delisted gpt-oss-120b-input
2026-08-27 requesty raised gpt-oss-120b-output 11.1%: 0.54 → 0.6 usd_per_mtok
2026-08-27 requesty raised gpt-oss-120b-input 11.1%: 0.135 → 0.15 usd_per_mtok
2026-08-22 openrouter raised openai-gpt-oss-120b-input 23.3%: 0.03 → 0.037 usd_per_mtok

The cheapest tracked gateway charges $0.03 per million input tokens and $0.14 per million output tokens (Glama). Prices refresh continually.

Glama for input tokens and Requesty for output tokens, among the 5 gateways we track.

Yes — the most expensive tracked gateway charges 1.7× the cheapest for input tokens. Identical model, identical tokens, different bill.