---
title: "OpenAI Introduces GPT-6.2 Astra and GPT-6.2 Sol: Smarter, More Accurate, and Cheaper Than Predecessors"
description: "OpenAI introduces GPT-6.2 Astra and GPT-6.2 Sol: Astra outperformed Claude Fable 5.2 by 5.5% at 5% of the competitor's cost, while Sol trails the flagship by just 0.5% at one-fifth the price."
date: 2026-11-18T01:36:42.460Z
lang: en
url: https://xab.info/en/posts/openai-gpt-6-2-astra-sol-new-ai-models
tags: [openai, gpt-6-2, artificial-intelligence, machine-learning, ai-models, claude]
author: "Roman V."
publisher: "XAB.info"
---

# OpenAI Introduces GPT-6.2 Astra and GPT-6.2 Sol: Smarter, More Accurate, and Cheaper Than Predecessors

![Two cute white robots working on laptops with a glowing blue OpenAI neon logo in the background](https://xab.info/media/2026/11/18/openai-gpt-6-2-astra-sol-novye-modeli-ii/openai-gpt-6-2-astra-sol-novye-modeli-ii-1.webp)

## Key Points

- OpenAI released GPT-6.2 Astra (professional tasks) and GPT-6.2 Sol (everyday tasks), both cheaper than predecessors
- Astra outperformed Claude Fable 5.2 by 5.5% in Agents' Last Exam at 5% of the competitor's cost
- In Terminal-Bench-Science 0.1, Astra scored 9% higher than its predecessor with a one-third cost reduction
- In AA-3D, Astra scored 1,318 Elo compared to Claude Fable 5.2's 1,271
- Sol lags behind Astra in FrontierCode 2.0 by only 0.5% at about one-fifth the cost

OpenAI has announced the release of two new artificial intelligence models — GPT-6.2 Astra and GPT-6.2 Sol. The former is focused on solving complex professional tasks, while the latter targets everyday user workflows. According to the company, both models have become smarter and more accurate than their predecessors, while their usage costs have not only remained stable but actually decreased.

### Professional Tasks: Astra vs. Competitors

The most noticeable improvements in GPT-6.2 Astra appear in professional document preparation, data analysis, and scientific research. In the Agents' Last Exam benchmark, which evaluates multi-step professional tasks, the new model outperformed GPT-6 Astra by 5.2% while consuming less than a third of the resources. Comparisons with Claude Fable 5.2 proved even more striking: GPT-6.2 Astra surpassed its competitor by 5.5% at a cost amounting to just 5% of Claude Fable 5.2's usage price. The internal GDPval v2 test recorded a 6.2% gain relative to GPT-6 Astra while reducing output token consumption by 22%.

### Scientific Research and Data Processing

OpenAI emphasizes that the Terminal-Bench-Science 0.1 benchmark confirms GPT-6.2 Astra's status as the top model for scientists. This test evaluates models' ability to perform scientific work controlled via command line. GPT-6.2 Astra scored 9% higher than its predecessor at a third lower usage cost. The model also claimed first place in the Vals scientific index, where the AI is tasked with interpreting real biological data and conducting experiments to understand simulated systems. Additionally, the company highlights improvements in presentation quality, financial tables, legal documents, and analytical reports: texts have become simpler, more accessible, and contain less filler.

### 3D Modeling and Engineering Tasks

GPT-6.2 Astra has achieved impressive successes in 3D modeling. The press release demonstrates this using the visualization of car chassis dynamics on a mountain road: the model created a 3D model, wrote a Python calculation block to simulate complex tire-suspension interaction dynamics, and then animated the results in Blender. In the AA-3D test, GPT-6.2 Astra scored 1,318 Elo points, outperforming Claude Fable 5.2 (1,271 points) and GPT-6.2 Sol (1,259 points). Increased calculation accuracy for individual parts and assemblies enables engineers to quickly test various structural combinations and select the most reliable solutions. The announcement also showcases the development of a remote-controlled car complete with blueprints, nets, and a bill of components for DIY assembly.

### GPT-6.2 Sol: Everyday Work Without Compromise

GPT-6.2 Sol is positioned as the primary model for everyday tasks. Compared to GPT-6.1 Sol, it is smarter and more efficient, allowing users to accomplish more tasks within the same subscription tier. The model supports multi-step agent workflows, writes clearly and naturally, and provides an intuitive interaction experience. Many OpenAI employees use GPT-6.2 Sol as their baseline model for both technical and non-technical tasks. In Cognition's FrontierCode 2.0 benchmark, GPT-6.2 Sol lags behind GPT-6.2 Astra by only 0.5% at roughly one-fifth of the cost (both models were tested on maximum settings).

### Web Design, Code, and Virtual Worlds

OpenAI also states that GPT-6.2 Astra brings deeper visual content to web design: the model selects layouts and images that give each website a unique look and structures elements according to their actual purpose. The new model writes cleaner code and creates interfaces that look and work better, especially for websites, games, and 3D design. Furthermore, Astra recreates virtual worlds, digital twins, and video games. According to several tests, both new OpenAI models approach or exceed the performance of other leading models at significantly lower task execution costs.

## Fact-Check & Sources

- [OpenAI представила две «жутко умные» модели ИИ — GPT-6.2 Astra для профи и GPT-6.2 Sol для повседневных дел](https://3dnews.ru/1151437) - Primary source. All benchmark figures (Agents' Last Exam, Terminal-Bench-Science 0.1, Vals, GDPval v2, AA-3D, FrontierCode 2.0) and model characteristics are taken from this text.
- [OpenAI представила две «жутко умные» модели ИИ — GPT-6.2 Astra для профи и GPT-6.2 Sol для повседневных дел](https://3dnews.ru/1151437/openai-predstavila-dve-gutko-umniemodeli-ii-gpt62-astra-dlya-profi-i-gpt62-sol-dlya-povsednevnih-del) - Duplicates source 1, contains no additional facts.
- [OpenAI представила GPT-6 Sol и Luna: новые модели стали вдвое дешевле своих предшественников](https://zn.ua/TECHNOLOGIES/openai-predstavila-gpt-6-sol-i-luna-novye-modeli-stali-vdvoe-deshevle-svoikh-predshestvennikov.html) - Describes an earlier release (GPT-6 Sol and GPT-6 Luna, September 2026) rather than GPT-6.2 Astra/Sol from the base. No direct contradictions exist — they refer to different model generations.
- [GPT-6 Sol и Luna: OpenAI приносит интеллект Astra в дешёвый сегмент](https://www.sostav.ru/blogs/292765/110578) - Confirms chronology: GPT-6 Astra was released earlier, followed by GPT-6 Sol/Luna (09.22.2026). GPT-6.2 Astra/Sol from the base is a later release. No contradictions with the base text.

## FAQ

### How does GPT-6.2 Astra differ from GPT-6.2 Sol?

GPT-6.2 Astra focuses on complex professional tasks: document preparation, data analysis, scientific research, and 3D modeling. GPT-6.2 Sol serves as the primary model for everyday work: multi-step agent workflows, text writing, and handling technical and non-technical tasks. Sol is about five times cheaper than Astra while trailing it by less than 0.5% in certain benchmarks.

### By how much does GPT-6.2 Astra outperform Claude Fable 5.2?

In the Agents' Last Exam benchmark, GPT-6.2 Astra beat Claude Fable 5.2 by 5.5% at a usage cost representing just 5% of Claude Fable 5.2's price. In the AA-3D test, Astra scored 1,318 Elo points against Claude Fable 5.2's 1,271 points.

### How much does it cost to use the new models?

OpenAI states that the usage cost of GPT-6.2 Astra and GPT-6.2 Sol has not increased compared to predecessors but rather decreased. Specific token prices are not listed in the press release, but it notes that Astra performs tasks using less than a third of the resources compared to GPT-6 Astra, while Sol operates at roughly one-fifth of Astra's cost.

### What 3D modeling capabilities does GPT-6.2 Astra demonstrate?

The press release shows Astra creating a 3D model of car chassis dynamics on a mountain road, writing a Python calculation block for tire-suspension interaction, and animating the results in Blender. It also demonstrates the development of a remote-controlled car with blueprints, nets, and a component list.