GPT-6 Astra is a proprietary multimodal reasoning model from OpenAI that accepts text and image input and produces text output. It is positioned as the company's flagship system for long-horizon end-to-end work spanning complex reasoning, software engineering, computer use, browsing, research and document creation. The model exposes a graduated reasoning effort control with low, medium, high, xhigh and max settings, and it accepts a change to that setting partway through a conversation rather than only at request time. It launches as a single tier with no smaller mini or nano variants, carries a context window of roughly 1.05 million tokens with a maximum output of 128,000 tokens, and reports a knowledge cutoff of April 30, 2026.
OpenAI reports evaluation results across agentic, scientific and security benchmarks, including 96.0% on GPQA Diamond, 64.6% on Terminal-Bench Science, 72.6% on OSWorld 2.0, and a perfect score on ExploitBench, along with near saturation on FrontierMath Tier 4 and ARC-AGI-3. The model supports computer use, structured outputs, streaming, programmatic tool calling, multi-agent orchestration, prompt caching and persisted reasoning, and it keeps earlier context windows searchable so it can recover requirements or tool outputs from previous turns. OpenAI describes Astra as the first of its models to cross the Critical cybersecurity capability threshold under its Preparedness Framework.
Drag and drop an image here, or click to browse
Model settings
Thinking level
Max output tokens
Default 65,536 · max 65,536
Sign in to adjust thinking and output length per run.
Results appear here. Add an image or pick an example to run GPT-6 Astra.
—
Usage
Past 30 DaysVision Evals is Roboflow's ground-truth benchmark: every model runs the same real-world samples across six vision tasks, and answers are scored against ground truth.
Evals updated September 5, 2026Pricing updated September 5, 2026
GPT-6 Astra averages 86.6% across the six Vision Evals tasks, ranking #1 of 53 models overall.
It leads the field in Object Detection and Reasoning.
It also places in the top three for Counting.
Its weakest relative showing is OCR, ranking #14 of 53 at 91.9%.
At $0.030 per sample it is the 51st cheapest of the 53 benchmarked models, and its average inference time of 6.7s per sample makes it the 13th fastest.
Field medians: Object Detection 53.9%, Counting 56.8%, Identification 84.4%, OCR 88.7%, Data Extraction 84.5%, Reasoning 54.1%.
| Task | Score | Field (0 to 100) | Rank | Cost / sample | Speed |
|---|---|---|---|---|---|
| Object Detection (low) | 82.1% ±0.8, Mean of 3 runs, range 81.0 to 82.7 | #1 of 53 | $0.050 | 10.96s | |
| Object Detection (high) | 83.6% ±0.8, Mean of 3 runs, range 82.8 to 84.5 | #1 of 17 | $0.101 | 31.66s | |
| Counting (low) | 80.2% ±1.4, Mean of 3 runs, range 78.4 to 81.1 | #3 of 53 | $0.018 | 4.39s | |
| Counting (high) | 81.1% ±1.4, Mean of 3 runs, range 79.7 to 82.4 | #2 of 17 | $0.028 | 7.80s | |
| Identification (low) | 89.6% ±1.6, Mean of 3 runs, range 87.5 to 90.6 | #14 of 53 | $0.013 | 2.47s | |
| Identification (high) | 92.7% ±1.6, Mean of 3 runs, range 90.6 to 93.8 | #6 of 17 | $0.014 | 3.09s | |
| OCR (low) | 91.9% ±0.2, Mean of 3 runs, range 91.6 to 92.1 | #14 of 53 | $0.031 | 7.02s | |
| OCR (high) | 91.5% ±0.2, Mean of 3 runs, range 91.3 to 91.7 | #5 of 17 | $0.089 | 25.89s | |
| Data Extraction (low) | 88.7% ±1.0, Mean of 3 runs, range 87.6 to 89.7 | #13 of 53 | $0.015 | 2.94s | |
| Data Extraction (high) | 91.1% ±1.0, Mean of 3 runs, range 89.7 to 91.8 | #6 of 17 | $0.018 | 4.18s | |
| Reasoning (low) | 87.2% ±1.0, Mean of 3 runs, range 86.1 to 88.1 | #1 of 53 | $0.016 | 3.91s | |
| Reasoning (high) | 91.2% ±0.3, Mean of 3 runs, range 90.7 to 91.4 | #1 of 39 | $0.021 | 5.51s |
Overall benchmark score against estimated cost per sample, on a log scale. Upper-left is the sweet spot: high quality at low cost.
52 models on the current benchmark · scores and efficiency pooled across all six tasks at low effort · GPT-6 Astra highlighted
GPT-6 Astra scores are the mean of 3 runs per task at both low and high effort · Methodology
View all Vision Evals →GPT-6 Astra costs $10.00 per 1M input tokens and $50.00 per 1M output tokens.
Pricing updated Sep 5, 2026
GPT-6 Astra is proprietary: the weights are not distributed, and the GPT-6 Astra license is the vendor's commercial terms of service that you accept when you call the API.
Vendor terms govern data retention, whether your inputs can be trained on, rate limits, and regional availability, and they can change with notice. Review them if you handle regulated or customer data.
Proprietary terms are set by the vendor rather than negotiated per project, and no open-source obligation attaches to your code. If you would rather deploy a model whose commercial license is included in your plan — on Roboflow Managed Cloud or a Self-Hosted Inference Server — Roboflow's licensing page lists the supported alternatives to GPT-6 Astra.
Do not hesitate to reach out with questions for your commercial project — our team will help you start solving business problems on the first call. See Roboflow commercial licensing for the models included in each plan.
Talk to salesThis model is proprietary. The author retains all rights, and use of the model is governed by their specific terms of service or license agreement.
Commercial use depends on the terms set by the model author. Most proprietary commercial models require a paid subscription, API key, or per-call billing. Check the provider’s pricing and terms-of-service for details.
License information is provided as a guide and is not legal advice.
Yes. GPT-6 Astra accepts image input and handles OCR, data extraction, object counting, identification, visual reasoning, and object detection. On Roboflow's Vision Evals its strongest task is Object Detection at 82.1% (#1 of 53 at low effort). You can test it on your own image in the demo above.
Yes, and it is one of the model's strongest vision skills: its transcriptions match the ground truth 91.9% on average (#14 of 53 at low effort) on Vision Evals OCR. Pulling specific fields out of documents (data extraction) scores 88.7%.
Yes. On Vision Evals, GPT-6 Astra scores 82.1% mAP@50 on object detection (#1 of 53 at low effort) and 80.2% judge-graded accuracy on object counting.
On our benchmark's task mix, GPT-6 Astra averages $0.03 per sample at $10.00 per 1M input and $50.00 per 1M output tokens (#51 of 53 on cost), with an average speed of 6.7s per sample across the benchmark. Actual cost depends on your images and prompts.
On the overall Vision Evals ranking, GPT-6 Astra sits #1 of 53 at 86.6%, just ahead of Gemini 3.5 Flash (86%). See the full side-by-side: GPT-6 Astra vs Gemini 3.5 Flash.