# GPT-6.2 Astra

> GPT-6.2 Astra was released by OpenAI on November 17, 2026, the same day as GPT-6.2 Sol, as the newer Astra after GPT-6 Astra of September 3. OpenAI's model page describes it as its most capable model for the hardest projects, from start to finish, able to build and revise software, use a computer and work with documents. OpenAI's GPT-6.2 guide suggests GPT-6.2 Astra for the most demanding tasks and GPT-6.2 Sol for complex work at lower cost.

Canonical: https://routemux.com/openai/gpt-6.2-astra
Provider: OpenAI

## Key facts

| Field | Value |
| --- | --- |
| Model ID | openai/gpt-6.2-astra |
| Official model ID | gpt-6.2-astra |
| Context window | 1.05M |
| Max output | 128K |
| Input modalities | text, image |
| Output modalities | text |
| Capabilities | Vision, Reasoning, Tool Calling, Coding, Cache, Paid Cache Write |
| Upstream release | 2026-11-17 |
| Availability | available |

## Pricing

- RouteMux input: $1.2 per 1M tokens
- RouteMux output: $6 per 1M tokens
- Official input: $8 per 1M tokens
- Official output: $40 per 1M tokens
- Cache read: $0.06 per 1M tokens
- Cache write: $1.5 per 1M tokens
- Saving vs official: 85%

Billing is prepaid wallet, charged per successful request only.

## API access

Base URL: https://api.routemux.com

Supported protocols:

- openai_chat
- openai_responses
- anthropic_messages

## What is GPT-6.2 Astra

GPT-6.2 Astra was released by OpenAI on November 17, 2026, the same day as GPT-6.2 Sol, as the newer Astra after GPT-6 Astra of September 3. OpenAI's model page describes it as its most capable model for the hardest projects, from start to finish, able to build and revise software, use a computer and work with documents. OpenAI's GPT-6.2 guide suggests GPT-6.2 Astra for the most demanding tasks and GPT-6.2 Sol for complex work at lower cost.

The specs: a 1.05M-token context window (up to 922K input), 128K max output, text and image in, text out, and a knowledge cutoff of May 18, 2026. Reasoning effort takes low, medium, high, xhigh or max, with medium as the default; OpenAI states that the API rejects requests that set reasoning effort to none. OpenAI's docs also list a Multi-agent beta in the Responses API that can delegate work to subagents. On RouteMux it supports tool calling and explicit prompt caching, and it is reachable over OpenAI Chat Completions, OpenAI Responses and Anthropic Messages.

What changes when you migrate: OpenAI says that starting with GPT-6.2, the API drops reasoning items when it detects a mid-conversation change to the conversation history (a prefix edit), a model downgrade, or a change of end user or organization, which can lower quality and use more tokens. OpenAI recommends keeping the prompt prefix unchanged, which preserves reasoning and keeps earlier input eligible for prompt caching. OpenAI also notes that 6.2 Astra writes more concisely and directly by default, so existing "be concise" instructions are worth revisiting, and that when intent is unclear it may propose a next step and wait for confirmation. GPT-6 Astra remains available on RouteMux if your prompts and evals are tuned against it.

At the standard tier (input up to 272K tokens) it bills $1.2 per 1M input tokens and $6 per 1M output tokens, with cached input at $0.06 per 1M tokens. OpenAI lists $8, $40 and $0.4 for the same three.

## Key capabilities

- **Built for the hardest projects** — OpenAI describes it as its most capable model for the hardest projects, from start to finish: building and revising software, using a computer and working with documents.
- **Reasoning from low to max** — reasoning_effort takes low, medium, high, xhigh or max, defaulting to medium; OpenAI states the API rejects none.
- **More concise by default** — OpenAI says 6.2 Astra writes more concisely and directly by default, and may propose a next step and wait for confirmation when intent is unclear.
- **Knowledge cutoff May 18, 2026** — 18 days later than GPT-6 Astra's April 30; the 1.05M context and 128K output match GPT-6 Astra.

## Other variants in this family

- GPT-6.2 Sol (`openai/gpt-6.2-sol`)
- GPT-6 Astra (`openai/gpt-6-astra`)
- GPT-6.1 Sol (`openai/gpt-6.1-sol`)
- GPT-6 Sol (`openai/gpt-6-sol`)
- GPT-6.1 Terra (`openai/gpt-6.1-terra`)
- GPT-6 Luna (`openai/gpt-6-luna`)

## Sources

- [OpenAI — GPT-6.2 Astra model page](https://developers.openai.com/api/docs/models/gpt-6.2-astra)
- [OpenAI — Using GPT-6.2 (model guide)](https://developers.openai.com/api/docs/guides/latest-model)
- [OpenAI — API changelog](https://developers.openai.com/api/docs/changelog)

_Content reviewed by the RouteMux team on 2026-11-18._
