Breaking

OpenAI Rolls Out GPT-6.1 Sol Ultrafast Across API, Codex and ChatGPT Work

October 8, 2026 at 14:27 EDT

Foundation Models
AI Agents
Software Dev & Coding

OpenAI has begun rolling out Ultrafast for GPT-6.1 Sol, adding a premium speed tier for its lower-cost near-Astra model in the API, Codex and ChatGPT Work. The move gives developers a faster option for agentic coding and professional workflows where waiting on generated tokens can slow iteration.

Oct. 8, 2026 | OpenAI

GPT-6.1 Sol gets Ultrafast

A premium speed tier is rolling out for GPT-6.1 Sol across the API, Codex and ChatGPT Work, aimed at faster agentic coding and professional workflows.

The main speed claim, drawn to scale

Standard

up to 8x

Ultrafast in Codex

The improvement refers to token generation speed, not guaranteed end-to-end task completion.

$12

API input per 1M tokens, short context

$60

API output per 1M tokens, short context

Default Build tier Ultrafast TPM

Purchased credits and Enterprise pay-as-you-go rate

Where it fits

API

Set `service_tier` to `ultrafast` with `gpt-6.1-sol`.

Codex

Use faster generation for coding loops where latency is the bottleneck.

ChatGPT Work

Available on Pro $500 and eligible Enterprise and Edu plans.

Why teams want it

Faster streaming can make long coding, document and agent workflows feel less blocked when output generation is the slow step.

What to watch

The tier costs more, depends on plan and workspace settings, and does not remove delays from tool calls or reviews.

Continue reading

The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.