OpenAI: gpt-oss-120b
Its listed blended standard token rate at OpenRouter rose 273.7% between two snapshots.
$0.070→$0.26
USD per 1M tokens · blended at 3 input : 1 output
Input$0.037 → $0.15+305.4%
Output$0.17 → $0.60+252.9%
Exact listing IDopenai/gpt-oss-120bOpenRouter
What it means for a workload
For a workload using 10M uncached input tokens and 2M output tokens, the listed token cost changes from $0.71 to $2.70. This excludes caching, batch discounts, long-context tiers, regional terms, latency, and reliability.
Historical context
LLMPI has 93 snapshot observations for this exact source listing since 2026-06-14. The observed blended range is $0.060–$0.26 per 1M tokens.
Evidence and scope
- Source
- OpenRouter
- Previous snapshot
- 2026-10-12T11:30:14+00:00
- Detected snapshot
- 2026-10-13T11:30:31+00:00
This is an observed change in the same source listing ID, not a claim that every endpoint or billing tier changed, or that this route is best on performance.