Roboflow
OpenAI

OpenAI: GPT-6 Luna

Not available to run yet

GPT-6 Luna is on the Vision Evals leaderboard. Running it in the Playground is not available yet. View evals

GPT-6 Luna Overview

GPT-6 Luna is the fast, cost-efficient tier of OpenAI's GPT-6 model family, sitting below GPT-6 Sol and the larger GPT-6 Astra model that opened the generation. It is a proprietary multimodal transformer that accepts text and image input and returns text, and it exposes an adjustable reasoning effort setting so the same model can run in a low-latency mode or spend additional inference compute on harder problems. OpenAI positions it for high-volume and latency-sensitive workloads such as conversational assistants, classification, and lightweight agentic pipelines, while noting that at higher reasoning effort it handles software engineering and computer-use tasks that previously required a Sol-tier model.

The model supports a context window of roughly 1,050,000 input tokens with a maximum output of 128,000 tokens, which allows long documents, extended agent traces, and large code repositories to be processed in a single request. OpenAI describes the GPT-6 generation as improving factual reliability and adopting a more concise communication style relative to the GPT-5.6 series, and attributes the efficiency of the Sol and Luna tiers to gains in caching and inference rather than to reduced capability. Architecture details, parameter counts, and training data are not published.

GPT-6 Luna Details & Performance

Details

Resources

Vision Tasks

CaptioningChart Question AnsweringClassificationDocument Question AnsweringImage TaggingMulti-Label ClassificationOCRVision LanguageVisual Question AnsweringObject Detection

Features

Foundation VisionLLMs with Vision CapabilitiesMultimodal Vision

Usage

Past 30 Days

Not available

Not in Playground

Performance

Avg. Latency

GPT-6 Luna Vision Evals

Vision 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 22, 2026Pricing updated September 22, 2026

Overall score#32 of 57
68.6%
Avg cost / sample#7 of 57
$0.0004
Avg speed / sample#33 of 57
11.27s
Avg tokens / sample
2.1K

Strengths and weaknesses

GPT-6 Luna averages 68.6% across the six Vision Evals tasks, ranking #32 of 57 models overall.

Its weakest relative showing is Data Extraction, ranking #54 of 57 at 68.0%.

At $0.0004 per sample it is the 7th cheapest of the 57 benchmarked models, and its average inference time of 11.3s per sample makes it the 33rd fastest.

Performance profile

Field medianGPT-6 Luna

Field medians: Object Detection 54.3%, Counting 61.7%, Identification 84.4%, OCR 88.7%, Data Extraction 84.5%, Reasoning 54.8%.

Results by task

TaskScoreField (0 to 100)RankCost / sampleSpeed
Object Detection (low)
56.8%
±1.9, Mean of 3 runs, range 54.8 to 58.5
#24 of 57$0.000615.29s
Object Detection (high)
64.1%
±0.5, Mean of 3 runs, range 63.6 to 64.5
#13 of 23$0.001626.66s
Counting (low)
65.8%
±1.4, Mean of 3 runs, range 64.9 to 67.6
#22 of 57$0.00038.88s
Counting (high)
64.4%
±2.0, Mean of 3 runs, range 62.2 to 66.2
#19 of 23$0.000613.31s
Identification (low)
81.3%
±0.0, Mean of 3 runs, range 81.3 to 81.3
#38 of 57$0.00026.30s
Identification (high)
80.2%
±1.6, Mean of 3 runs, range 78.1 to 81.3
#22 of 23$0.00037.30s
OCR (low)
87.9%
±0.6, Mean of 3 runs, range 87.2 to 88.3
#36 of 57$0.00058.59s
OCR (high)
88.5%
±0.6, Mean of 3 runs, range 87.9 to 89.2
#21 of 23$0.001422.29s
Data Extraction (low)
68.0%
±3.1, Mean of 3 runs, range 65.0 to 71.1
#54 of 57$0.00027.19s
Data Extraction (high)
66.7%
±0.5, Mean of 3 runs, range 66.0 to 67.0
#23 of 23$0.00049.59s
Reasoning (low)
52.1%
±2.0, Mean of 3 runs, range 49.7 to 53.6
#32 of 57$0.000310.13s
Reasoning (high)
60.7%
±1.7, Mean of 3 runs, range 58.9 to 62.3
#34 of 43$0.000612.74s
  • Thinking longer helps: 7.4 points higher on object detection at high effort for 2.6x the cost and 1.7x the latency.
  • Thinking longer does not help: 1.4 points lower on counting at high effort for 2x the cost and 1.5x the latency.
  • Thinking longer does not help: 1 points lower on identification at high effort for 1.6x the cost and 1.2x the latency.
  • Thinking longer helps: 0.6 points higher on ocr at high effort for 2.7x the cost and 2.6x the latency.
  • Thinking longer does not help: 1.4 points lower on data extraction at high effort for 1.6x the cost and 1.3x the latency.
  • Thinking longer helps: 8.6 points higher on reasoning at high effort for 2.1x the cost and 1.3x the latency.

Price vs. performance

Score vs. cost

Overall benchmark score against estimated cost per sample, on a log scale. Upper-left is the sweet spot: high quality at low cost.

56 models on the current benchmark · scores and efficiency pooled across all six tasks at low effort · GPT-6 Luna highlighted

GPT-6 Luna scores are the mean of 3 runs per task at both low and high effort · Methodology

View all Vision Evals →

GPT-6 Luna Pricing

GPT-6 Luna costs $0.100 per 1M input tokens and $0.500 per 1M output tokens.

Input$0.100 / 1M tokens
Output$0.500 / 1M tokens
Cached input$0.010 / 1M tokens

Pricing updated Sep 22, 2026

Other OpenAI GPT Nano models

Other versions in the same family as GPT-6 Luna.

GPT-6 Luna License

Proprietary

GPT-6 Luna is proprietary: the weights are not distributed, and the GPT-6 Luna license is the vendor's commercial terms of service that you accept when you call the API.

Commercial use
Permitted under the vendor terms, typically metered per token or per request, with the vendor usage policy applying to your inputs and outputs.
Modification
Not available. GPT-6 Luna weights are closed, so you can configure prompts and use vendor-hosted fine-tuning where it is offered, but you cannot modify the model itself.
Redistribution
Not permitted. You cannot self-host or resell the model; you build on the hosted API instead.

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.

Do I need a commercial license for GPT-6 Luna?

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 Luna.

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 sales

This 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.

Frequently Asked Questions About GPT-6 Luna Vision

Yes. GPT-6 Luna 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 Counting at 65.8% (#22 of 57 at low effort).

Yes. its transcriptions match the ground truth 87.9% on average (#36 of 57 at low effort) on Vision Evals OCR. Pulling specific fields out of documents (data extraction) scores 68%.

It's serviceable. On Vision Evals, GPT-6 Luna scores 56.8% mAP@50 on object detection (#24 of 57 at low effort) and 65.8% judge-graded accuracy on object counting.

On our benchmark's task mix, GPT-6 Luna averages $0.0004 per sample at $0.10 per 1M input and $0.50 per 1M output tokens (#7 of 57 on cost), with an average speed of 11.3s per sample across the benchmark. Actual cost depends on your images and prompts.

On the overall Vision Evals ranking, GPT-6 Luna sits #32 of 57 at 68.6%, just behind Claude Opus 4.8 (68.7%) and just ahead of Qwen3.7 Plus (67.4%). See the full side-by-side: GPT-6 Luna vs Claude Opus 4.8.