OpenAI: GPT-5.6 Luna
GPT-5.6 Luna is an OpenAI model that accepts text, image, and file inputs and produces text output. Its context window holds 1,050,000 tokens, and a single response can run up to 128,000 tokens, which suits long documents, large codebases, or extended multi-file sessions. It supports tool calling and reasoning; the available data does not confirm structured output support or whether weights are open. The pricing is $0.20 per million input tokens and $1.20 per million output tokens, which places it at the low end for a model with this context size. Its blended benchmark score is 89.4 across three benchmarks. The split is uneven: it scores 100.0 on the coding index, 84.5 on the general AI index, and 75.2 on the agentic index. Teams running high-volume coding or long-context analysis on a budget should shortlist it. Agent-heavy workloads should be tested against alternatives first, since three benchmarks is limited coverage.
Benchmark results
Independent, published benchmarks. Blended score 89.4 across 3 benchmarks, last refreshed 2026-11-22. How scoring works →
| Benchmark | Measures | Score |
|---|---|---|
| AI Index | broad capability composite | 84.5 |
| AI Index Coding | software engineering tasks | 100.0 |
| AI Index Agentic | multi-step tool-using tasks | 75.2 |
- Model ID
- openai/gpt-5.6-luna
- Vendor
- openai
- Released
- July 2026
- Tokenizer
- GPT
- Input Modalities
- file, image, text
- Output Modalities
- text
- Max Output
- 128,000 tokens
- Tool Calling
- ✓ supported
- Structured Output
- ✓ supported
- Reasoning Mode
- ✓ supported
- Vision
- ✓ accepts images
- Audio
- no
- Moderated
- yes
What it costs in practice
Computed from the current $0.20/M input and $1.20/M output rates. Run your own numbers →
| Job | Tokens | Cost |
|---|---|---|
| Summarize a 50-page report | 30k in / 1.5k out | under $0.01 |
| Classify 1,000 customer emails | 500k in / 50k out | $0.16 |
| A month of a busy support chatbot | 5M in / 2M out | $3.40 |
Price & spec history
Tracked daily by PicksByModel since 2026-07-17.
| Date | Input /M | Output /M | Context |
|---|---|---|---|
| 2026-08-18 | $0.20 | $1.20 | 1,050,000 |
| 2026-07-31 | $0.10 | $0.60 | 1,050,000 |
| 2026-07-28 | $0.50 | $3.00 | 1,050,000 |
| 2026-07-17 | $1.00 | $6.00 | 1,050,000 |
Strong choice for
JSON Extraction
Bulk Data Labeling
Short-Form Summarization
Email Drafting
Language Learning
Customer Support
Category rankings
Where OpenAI: GPT-5.6 Luna places across the 49 categories it ranks in. How we rank →
| # | Category | Score |
|---|---|---|
| #1 | Image CaptioningVision · of 25 ranked | 121 |
| #2 | Bulk Data LabelingData · of 25 ranked | 135 |
| #2 | Customer SupportBusiness · of 25 ranked | 135 |
| #2 | Dataset AnnotationResearch · of 25 ranked | 145 |
| #3 | JSON ExtractionData · of 25 ranked | 149 |
| #3 | Short-Form SummarizationWriting · of 25 ranked | 132 |
| #3 | Email DraftingWriting · of 25 ranked | 128 |
| #3 | Language LearningEducation · of 25 ranked | 128 |
| #3 | Chat CompanionPersonal · of 25 ranked | 132 |
| #3 | Trivia & General KnowledgePersonal · of 25 ranked | 122 |
| #5 | Sales / Cold EmailBusiness · of 25 ranked | 124 |
| #5 | Job Application DraftingBusiness · of 25 ranked | 124 |
| #5 | Journaling HelperPersonal · of 25 ranked | 124 |
| #5 | Recipe GenerationPersonal · of 25 ranked | 124 |
| #9 | Code CompletionCode · of 25 ranked | 133 |
| #9 | Resume WritingBusiness · of 25 ranked | 118 |
| #10 | Regex WritingCode · of 25 ranked | 136 |
| #10 | Language TranslationWriting · of 25 ranked | 136 |
| #10 | Tutoring for KidsPersonal · of 25 ranked | 136 |
| #13 | SOP / Process DocsBusiness · of 25 ranked | 137 |
| #13 | Function / Tool CallingAgents · of 25 ranked | 154 |
| #15 | Code DocumentationCode · of 25 ranked | 143 |
| #15 | Marketing CopyWriting · of 25 ranked | 130 |
| #15 | Standardized Test PrepEducation · of 25 ranked | 130 |
| #15 | Meeting NotesBusiness · of 25 ranked | 152 |
| #17 | CSV / Spreadsheet CleanupData · of 25 ranked | 154 |
| #17 | Diagram ExtractionVision · of 25 ranked | 149 |
| #17 | Screenshot DebuggingVision · of 25 ranked | 142 |
| #17 | Browser AutomationAgents · of 25 ranked | 171 |
| #18 | SQL GenerationCode · of 25 ranked | 171 |
| #19 | Frontend Component DesignCode · of 25 ranked | 146 |
| #19 | OCR / Document ParsingData · of 25 ranked | 146 |
| #19 | Table Extraction from PDFsData · of 25 ranked | 146 |
| #19 | Blog Post WritingWriting · of 25 ranked | 140 |
| #19 | Chart & Graph ReadingVision · of 25 ranked | 152 |
| #19 | Trip PlanningPersonal · of 25 ranked | 140 |
| #20 | Data AnalysisData · of 25 ranked | 170 |
| #21 | Fitness CoachingPersonal · of 25 ranked | 134 |
| #22 | Bug FixingCode · of 25 ranked | 179 |
| #22 | Financial AnalysisProfessional · of 25 ranked | 158 |
| #22 | Math TutoringEducation · of 25 ranked | 147 |
| #22 | Character RoleplayPersonal · of 25 ranked | 150 |
| #22 | Scientific CodingResearch · of 25 ranked | 179 |
| #22 | Experiment DesignResearch · of 25 ranked | 156 |
| #23 | Long-Document SummarizationWriting · of 25 ranked | 159 |
| #23 | Agent WorkflowsAgents · of 25 ranked | 191 |
| #23 | Coding AgentsAgents · of 25 ranked | 191 |
| #25 | Unit Test GenerationCode · of 25 ranked | 154 |
| #25 | CI/CD PipelinesCode · of 25 ranked | 144 |
Similar models
OpenAI: GPT-6.2 Sol Pro
OpenAI: GPT-6.2 Sol Pro (batch)
OpenAI: GPT-6.2 Sol
OpenAI: GPT-6.2 Sol (batch)
OpenAI: GPT-6.1 Terra Pro
OpenAI: GPT-6.1 Terra Pro (batch)
Quick answers
- How much does OpenAI: GPT-5.6 Luna cost?
- $0.20 per million input tokens and $1.20 per million output tokens.
- What is OpenAI: GPT-5.6 Luna's context window?
- 1,050,000 tokens, roughly 1,575 pages of text in a single request.
- Does OpenAI: GPT-5.6 Luna support tool calling?
- It supports tool calling, structured output, a reasoning mode.
- Can OpenAI: GPT-5.6 Luna process images?
- Yes, it accepts image input.