GPT-5.6 Luna
OpenAI · released Jul 9, 2026
- Type
- Proprietary
- Input
- $1.00
- Output
- $6.00
- Cached
- Not published
List price · per 1M tokens · OpenAI at 1.1M context · source ↗ · a reseller below undercuts it; the table carries the spread
Our take
Written Aug 3, 2026GPT-5.6 Luna is OpenAI's flagship text model that can handle over one million tokens in a single request, released in July 2026. It accepts text, images and files, and is available across a wide price spectrum from budget to premium tiers depending on which host you choose.
Pick this for long-document analysis at full one-million-token scale, or for coding and web development where its measured skill is strongest. Choose the budget tier if cost matters more than the last few tokens per second, or the premium tier if throughput is critical. Skip it if you need creative writing at peak quality, want to self-host or fine-tune, or need verified speed data on the most expensive tiers.
The case for it
- Extremely large request limit: over one million tokens, rare among hosted models.
- Coding and web development is its strongest measured skill, with Arena scores holding steady across multiple evaluation dates.
- Maths and hard-prompt performance is unusually stable, shifting by under four points between dates — less volatile than typical leaderboard movement.
- Wide price range lets you trade cost for speed across seven tracked hosts.
The case against it
- Creative writing is its weakest measured skill, more than thirty points below its own overall chat score.
- No measured tokens per second for the top two price tiers, so throughput is unverified there.
- Weights are proprietary with no local or fine-tuning option.
How good is it?
IntelligencePuzzles, maths, exam questions
Arena Text (overall)34th of 143 · 1450.4via xHigh
CodingWriting and fixing code on its own
Arena Coding33rd of 143 · 1501.4via xHigh
AgenticPlanning, calling tools, staying on task
Arena Agent (IPS)13th of 36 · 0.035via xHigh
WritingWe do not rate this
Two boards come close and neither tests writing: Arena Creative Writing asks people which of two replies they prefer, and LiveBench Language tests whether a model understood a passage. So we show where GPT-5.6 Luna placed and give it no mark out of five.
These tests check whether a model follows instructions — a precondition for all the work above, but not a measure of how well that work is done, which is why they get no rating.
Every published score for this model16 scoresEvery figure we hold, from 16 boards, with who ran it and a link to the source — including the boards no rating above is built on.
Or rent it from someone else
Why this differs from the header. The strip above quotes OpenAI's own list price. This is the cheapest live offer at the widest standard context we hold, whoever is serving it — a reseller undercutting a lab is ordinary commerce, not an error.
- per 1M tokens
- $0.22 in / $1.32 out
- Context served
- 1.1M
- Throughput
- ~43 tok/s
| Provider | In / out per 1M tokens | Context | Throughput | Trains on prompts | Logs prompts | Zero retention |
|---|---|---|---|---|---|---|
| Amazon Bedrockus-east-1 | $0.22 / $1.32 | 1.1M | 43 tok/s | No | No | Unknown |
| OpenAIflex tier | $0.25 / $1.50 | 1.1M | 44 tok/s | No | Yesunknown period | Unknown |
| OpenRouter | $0.50 / $3.00 | 1.1M | not measured | Unknown | Unknown | Unknown |
| OpenAI | $1.00 / $6.00 | 1.1M | not measured | No | Yesunknown period | Unknown |
| Microsoft Azure AI | $1.00 / $6.00 | 1.1M | 55 tok/s | No | No | Confirmed |
| Microsoft Azure AIeu | $1.10 / $6.60 | 1.1M | 125 tok/s | No | No | Confirmed |
| OpenAIpriority tier | $2.00 / $12.00 | 1.1M | not measured | No | Yesunknown period | Unknown |
OpenRouter
Microsoft Azure AIAcross the 7 listings we hold: 6 say they do not train on prompts, 0 say they do and 1 do not say. 2 appear in the zero-retention registry we check; the rest are unknown to us rather than confirmed either way.
What each host's API supports
From the parameter list each endpoint publishes. Streaming is omitted: nothing we hold reports it, for any model.
- ✓
- Supported
- ✗
- Not supported
- Not published
- host gave no parameter list
| Provider | Tool calling | JSON output | Strict schema |
|---|---|---|---|
| Amazon Bedrockus-east-1 | ✓ | ✗ | ✗ |
| OpenAIflex | ✓ | ✓ | ✓ |
| OpenRouter | ✓ | ✓ | ✓ |
| OpenAI | ✓ | ✓ | ✓ |
| Microsoft Azure AI | ✓ | ✓ | ✓ |
| Microsoft Azure AIeu | ✓ | ✓ | ✓ |
| OpenAIpriority | ✓ | ✓ | ✓ |
Tool calling: 7 of 7 listings say yes. JSON output: 6 of 7 listings say yes, 1 says no. Strict schema: 6 of 7 listings say yes, 1 says no.
Models people weigh against GPT-5.6 Luna
When we formed this view
Dates behind this page
Prices last checked 9d ago
What we do not know about this model yet
- Nothing we hold says whether an endpoint streams, so we do not show it either way.
- 1 of 7 listings do not say whether they train on prompts.
Licence and identifiers
What the licence allowsWe hold no licence record for this model. Inside are the identifiers you need to pull it — its Hugging Face repo where we have one, our slug and a machine-readable card.
Licence
Commercial API terms. We hold no licence record for this model, so there is nothing to summarise here.
Identifiers
- Modality record
- text+image+file->text
- Catalogue slug
- openai-gpt-5-6-luna