Ultra Fast
A high-speed inference service tier introduced by OpenAI, offering low-latency and high-throughput model invocation. It targets production environments where response speed is a critical requirement.
Timeline (last 90 days)
Ultra Fast mode outputs at approximately 8 times the speed of standard mode, is available only to the Pro plan at $500 per month, and consumes approximately 6 times the usage quota.
With Ultra Fast enabled, Astra's input price spikes to $60 per million tokens and output reaches as high as $300, and the $500 plan's weekly credits can be burned through in approximately 2.1 hours under Ultra Fast.
DevJ launched an Ultra Fast express lane priced at $1,500/month, targeting users with extreme demands for speed and throughput
The Ultra-Fast speed tier in Codex generates 300 tokens per second, is priced at 6 times the standard rate, and requires a $500 per month subscription.
OpenAI introduced the Ultra Fast service tier: up to 8x faster in Codex with 300 tokens per second generated, and up to 6x faster in the API.
The Ultra Fast tier for GPT-6 Astra is now open via API, priced at $60 per million tokens for short-context input and $300 per million tokens for output.
All Facts (6)
Ultra Fast mode outputs at approximately 8 times the speed of standard mode, is available only to the Pro plan at $500 per month, and consumes approximately 6 times the usage quota.
50%UnverifiedWith Ultra Fast enabled, Astra's input price spikes to $60 per million tokens and output reaches as high as $300, and the $500 plan's weekly credits can be burned through in approximately 2.1 hours under Ultra Fast.
50%UnverifiedThe Ultra Fast tier for GPT-6 Astra is now open via API, priced at $60 per million tokens for short-context input and $300 per million tokens for output.
50%UnverifiedDevJ launched an Ultra Fast express lane priced at $1,500/month, targeting users with extreme demands for speed and throughput
50%UnverifiedThe Ultra-Fast speed tier in Codex generates 300 tokens per second, is priced at 6 times the standard rate, and requires a $500 per month subscription.
50%UnverifiedOpenAI introduced the Ultra Fast service tier: up to 8x faster in Codex with 300 tokens per second generated, and up to 6x faster in the API.
50%