GPT-6.2 Astra
Latest GPT-6.2 AstraA frontier reasoning model for coding, computer use, and professional work such as slides, spreadsheets, and documents.
- Strongest at
- Writing Top 1% of the past year's models · one benchmark so far
- Price per 1M tokens
- $8 / $40 input / output · OpenAI
- Context window
- 1.05M tokens in · 128K out
- Since the last one
- 74 days after GPT-6 Astra (Fast)
Capability profile
Scored against 92 models released in the past year. 50 is the average model; further out is better.
Where these scores come from
Each capability combines the published benchmarks listed here, rescaled so 50 is the reference average and 15 points is one standard deviation. A hollow dot rests on a single benchmark.
- Coding MirrorCode
- Writing Lech Mazur Writing comparison
- Reading & documents GDP.pdf (Surge AI), Vision Evals: OCR & extraction (low effort)
- Reasoning SimpleBench, ARC-AGI-2, Chess Puzzles, Mystery Game Puzzles
- Math OTIS Mock AIME 2024-2025, FrontierMath-Tiers-1-3-v2-Private, FrontierMath-Tier-4-v2-Private
- Science GPQA diamond
- Visual reasoning Vision Evals (low effort), Blueprint-Bench 2, GDP.pdf (Surge AI)
- Computer use APEX-Agents, AutomationBench 1.0.6
Price against capability
The shaded corner holds anything better and cheaper. Prices blend input and output 3 to 1.
Nothing from the past year scores higher on Writing for less.
4 models from the past year score higher on Coding for less: Claude Opus 5.5, GPT-6.1 Terra, Claude Sonnet 5.5 and 1 more.
6 models from the past year score higher on Reading & documents for less: Gemini 4 Carbon, Gemini 4 Argon, Kimi K3.1 Pro and 3 more.
1 model from the past year scores higher on Reasoning for less: Gemini 4 Neon.
3 models from the past year score higher on Math for less: Kimi K3.1 Pro, GPT-6.1 Sol, Claude Opus 5.5.
8 models from the past year score higher on Science for less: Claude Opus 5.5, Claude Opus 5, Claude Sonnet 5.5 and 5 more.
5 models from the past year score higher on Visual reasoning for less: Kimi K3.1 Pro, Gemini 4 Carbon, Gemini 4 Neon and 2 more.
4 models from the past year score higher on Computer use for less: Gemini 4 Neon, Gemini 4 Carbon, Gemini 4 Argon and 1 more.
OpenAI $8 / $40 · OpenRouter $8 / $40 · Cached input $0.4
Since GPT-6 Astra (Fast)
3 releases in the GPT-6.2 Astra line.
Epoch benchmark boards
Best result on each board it has run. The tick marks the leader.
7 more boards
OpenAI's past year
26 release days in the past year, about one every 14 days. The next is expected around Nov 28, 2026. Forecasts ↗
Specifications
Specifications and prices from Models.dev; availability and limits may differ by provider. Benchmarks from Epoch AI (CC BY 4.0), LiveBench and the other sources named above.