Cheaper overall
GPT-6 Luna
57% lower blended rate
Higher benchmark scores
GPT-6 Luna
leads on 7 of 8 figures
Bigger context window
GPT-6 Luna
1,050,000 vs 400,000 tokens
Cheaper cached input
GPT-6 Luna
$0.01 vs $0.02, 50% less
Where each one wins
GPT-6 Luna
- Cheaper input tokens $0.10 vs $0.20, 50% less
- Cheaper output tokens $0.50 vs $1.25, 60% less
- Cheaper cached input $0.01 vs $0.02, 50% less
- Cheaper on all four workloads
- Larger context window 1,050,000 tokens
- Higher overall intelligence (Intelligence Index) 38.1 vs 20.7
- Higher expert-exam performance (Humanity's Last Exam) 38.5 vs 28.3
- Higher long-context reasoning (AA-LCR) 83.3 vs 76.7
- Ahead on 4 more benchmark figures
- Offers Priority pricing not listed for GPT-5.4 Nano
- Not scheduled for retirement GPT-5.4 Nano retires 1 Apr 2027
GPT-5.4 Nano
- Higher answer reliability (Omniscience Non-Hallucination) 48.9 vs 23.3
What four workloads cost
One run, and the same run a thousand times.
Caching is what decides it: over twenty calls GPT-6 Luna costs $0.0170 against $0.0375, a difference of $20.50 across a thousand loops.
| Workload | GPT-6 Luna | GPT-5.4 Nano | ×1,000 | Difference |
|---|---|---|---|---|
| Chat turn 1,000 in / 500 out | $0.0004 | $0.0008 | $0.3500 / $0.8250 | GPT-6 Luna is 58% cheaper |
| RAG answer 10,000 in / 800 out | $0.0014 | $0.0030 | $1.40 / $3.00 | GPT-6 Luna is 53% cheaper |
| Agent loop 32,000 in / 700 out × 20 calls | $0.0170 | $0.0375 | $17.00 / $37.50 | GPT-6 Luna is 55% cheaper |
| Long document 400,000 in / 3,000 out | $0.0823 | $0.0838 | $82.25 / $83.75 | GPT-6 Luna is 2% cheaper |
Price per 1M tokens
Above 272,000 input tokens the rates change, and the new rate applies to the whole request, not just the tokens past that point: GPT-6 Luna's input goes from $0.1000 to $0.2000. It does not change which of the two is cheaper.
| Context band | Price | GPT-6 Luna | GPT-5.4 Nano | Difference |
|---|---|---|---|---|
| Prompts up to 272,000 tokens | Input | $0.10 | $0.20 | GPT-6 Luna is 50% cheaper |
| Prompts up to 272,000 tokens | Cached input | $0.01 | $0.02 | GPT-6 Luna is 50% cheaper |
| Prompts up to 272,000 tokens | Output | $0.50 | $1.25 | GPT-6 Luna is 60% cheaper |
| Prompts over 272,000 tokens | Input | $0.20 | $0.20 | Same |
| Prompts over 272,000 tokens | Cached input | $0.02 | $0.02 | Same |
| Prompts over 272,000 tokens | Output | $0.75 | $1.25 | GPT-6 Luna is 40% cheaper |
Blended: GPT-6 Luna $0.20, GPT-5.4 Nano $0.4625 per 1M — one rate at 3:1 input to output, for comparing two models at a glance.
Both priced by OpenAI. Standard tier, pay-as-you-go.
Other pricing tiers
Both models offer Batch and Flex, worth up to 50% off their standard rates, and the cheaper model is the same one on every tier — so the tier you buy changes the bill rather than the choice. Only GPT-6 Luna lists Priority; GPT-5.4 Nano does not offer it.
Every tier either model sells is here, so this is the whole price sheet. A saving is measured against that model's own standard rate — how we read tiers.
| Tier | GPT-6 Luna | GPT-5.4 Nano | Cheaper |
|---|---|---|---|
| Batch | $0.05 in / $0.25 out 50% off Standard | $0.10 in / $0.625 out 50% off Standard | GPT-6 Luna |
| Flex | $0.05 in / $0.25 out 50% off Standard | $0.10 in / $0.625 out 50% off Standard | GPT-6 Luna |
| Priority only on GPT-6 Luna | $0.20 in / $1.00 out | Not offered | — |
Benchmarks
They share 8 figures: GPT-6 Luna leads on 7, GPT-5.4 Nano on 1. The widest gap is 25.6 points, on answer reliability (Omniscience Non-Hallucination).
Overall intelligence
Intelligence Index — Overall intelligence across reasoning, knowledge & math evals
GPT-6 Luna ahead by 17.4
Expert-exam performance
Humanity's Last Exam — Expert-level questions across many academic domains
GPT-6 Luna ahead by 10.2
Long-context reasoning
AA-LCR — Long-context reasoning across large inputs
GPT-6 Luna ahead by 6.6
Scientific coding
SciCode — Research-level scientific coding tasks
GPT-6 Luna ahead by 7.4
Physics research reasoning
CritPt — Unpublished physics research reasoning problems
GPT-6 Luna ahead by 10.1
Real-world task performance
GDPval — Economically valuable, real-world knowledge work
GPT-6 Luna ahead by 24.1
Factual accuracy
Omniscience Accuracy — Breadth of factual knowledge across domains
GPT-6 Luna ahead by 18.5
Answer reliability
Omniscience Non-Hallucination — How reliably the model avoids fabricated answers
GPT-5.4 Nano ahead by 25.6
Each figure is that model's best published run, with the effort level named beside it — where these scores come from.
Specifications
GPT-6 Luna holds 650,000 more input tokens in one request. GPT-5.4 Nano retires on 1 Apr 2027, so anything built on it needs a move planned.
| Specification | GPT-6 Luna | GPT-5.4 Nano |
|---|---|---|
| Context window | 1,050,000 tokens | 400,000 tokens |
| Max output | 128,000 tokens | 128,000 tokens |
| Takes in | Text, image, file | Text, image, file |
| Puts out | Text | Text |
| Knowledge cutoff | 18 May 2026 | 31 Aug 2025 |
| Released | 22 Sep 2026 | 17 Mar 2026 |
| Status | Active | Legacy |
| Retires | — | 1 Apr 2027 |
| Sold by | OpenAI | OpenAI |
FAQs
- Is GPT-6 Luna cheaper than GPT-5.4 Nano?
- Yes. A 1,000-token prompt with a 500-token reply costs $0.0004 on GPT-6 Luna and $0.0008 on GPT-5.4 Nano, and the same model is cheaper on every workload on this page.
- How much do GPT-6 Luna and GPT-5.4 Nano cost per 1M tokens?
- GPT-6 Luna costs $0.10 for input and $0.50 for output. GPT-5.4 Nano costs $0.20 and $1.25. Standard pay-as-you-go rates, as of 22 Sep 2026.
- Which is better, GPT-6 Luna or GPT-5.4 Nano?
- On published benchmarks, GPT-6 Luna. It leads on 7 of the 8 figures both models report, including overall intelligence (Intelligence Index), where it scores 38.1 against 20.7.
- Is Batch pricing cheaper for GPT-6 Luna or GPT-5.4 Nano?
- Batch input costs $0.05 on GPT-6 Luna and $0.10 on GPT-5.4 Nano, 50% below the standard rate on both.
- Does GPT-6 Luna or GPT-5.4 Nano have a bigger context window?
- GPT-6 Luna, at 1,050,000 tokens against 400,000. That is 650,000 more input tokens in a single request.