Ultrafast
OpenAI's premium speed tier delivering up to 8x faster token generation (300 tokens/sec) in Codex and up to 6x faster in the API.
Overview
OpenAI launched Ultrafast, a premium inference speed tier, at its September 29, 2026 DevDay. Ultrafast offers up to 8x faster token generation in Codex (up to 300 tokens per second) and up to 6x faster in the API. It is live today for GPT-6 Astra in the API, and in ChatGPT Work and Codex on the Pro 500 and Enterprise plans, with GPT-6.1 Sol Ultrafast support coming soon. The new $500/month Pro 500 ChatGPT plan gives roughly 25x the usage of ChatGPT Plus and removes the previous 5-hour usage window. API access to Ultrafast carries a steep markup, priced around $60 per million input tokens and $300 per million output tokens, roughly 6x GPT-6 Astra's standard rate.
This is a sourced directory entry. Features and capabilities are reported in the linked sources; this entry does not document independent Aymorphic testing.
Key features
- Up to 8x faster token generation in Codex, reaching up to 300 tokens/second
- Up to 6x faster token generation in the API
- Live today for GPT-6 Astra; GPT-6.1 Sol Ultrafast coming soon
- Available in ChatGPT Work and Codex on the new Pro 500 plan ($500/month) and Enterprise plans
- API pricing roughly 6x standard GPT-6 Astra rates (~$60/M input, ~$300/M output tokens)
Use cases
- Latency-sensitive agentic coding sessions in Codex
- High-throughput API workloads willing to pay a premium for speed
- Enterprise and Pro 500 subscribers needing maximum interactive responsiveness
Pricing
API: ~$60/million input tokens, ~$300/million output tokens (~6x standard GPT-6 Astra pricing). ChatGPT: included on the $500/month Pro 500 plan and Enterprise plans.
Pricing reflects the source record and may change. Confirm current terms with the provider.
Dates and sources
- Released
- 29 Sept 2026
- Record updated
- 3 Oct 2026
- Record last checked
- 3 Oct 2026