Models, measured side by side
GPT-6 Luna vs Phi-4
Compare GPT-6 Luna, Phi-4 by token pricing, context length, benchmark results, speed and tool support.
Monthly workload
Monthly cost breakdown
InputOutput
GPT-6 Luna
$4.50 / mo
Phi-4
$2.10 / mo
Performance comparison
GPT-6 Luna
38.1
Phi-4
5.9
Overview
Pricing & capacity
| Pricing & capacity | GPT-6 Luna | Phi-4 |
|---|---|---|
| Input / 1M tokens | $0.10 | $0.07 |
| Output / 1M tokens | $0.50 | $0.14 |
| Cached input / 1M | $0.01 | Not listed |
| Monthly cost | $4.50 | $2.10 |
| Context window | 1.1M | 16K |
| Maximum output | 128K | 15K |
Input / 1M tokens
- GPT-6 Luna
- $0.10
- Phi-4
- $0.07
Output / 1M tokens
- GPT-6 Luna
- $0.50
- Phi-4
- $0.14
Cached input / 1M
- GPT-6 Luna
- $0.01
- Phi-4
- Not listed
Monthly cost
- GPT-6 Luna
- $4.50
- Phi-4
- $2.10
Context window
- GPT-6 Luna
- 1.1M
- Phi-4
- 16K
Maximum output
- GPT-6 Luna
- 128K
- Phi-4
- 15K
Benchmarks & performance
| Benchmarks & performance | GPT-6 Luna | Phi-4 |
|---|---|---|
| Arena Elo | 1,211 | 1,123 |
| MMLU-Pro | 75.4% | 60.7% |
| Intelligence Index | 38.1 | 5.9 |
| GPQA | Not listed | 57.5% |
| Humanity’s Last Exam | 38.5% | 3.8% |
| SciCode | 54.6% | Not listed |
| Output speed | 141 tokens/s | 39 tokens/s |
| Time to first token | 95.97 s | 2.59 s |
| DesignArena rating | 1,277 | Not listed |
| DesignArena win rate | 48.5% | Not listed |
Arena Elo
- GPT-6 Luna
- 1,211
- Phi-4
- 1,123
MMLU-Pro
- GPT-6 Luna
- 75.4%
- Phi-4
- 60.7%
Intelligence Index
- GPT-6 Luna
- 38.1
- Phi-4
- 5.9
GPQA
- GPT-6 Luna
- Not listed
- Phi-4
- 57.5%
Humanity’s Last Exam
- GPT-6 Luna
- 38.5%
- Phi-4
- 3.8%
SciCode
- GPT-6 Luna
- 54.6%
- Phi-4
- Not listed
Output speed
- GPT-6 Luna
- 141 tokens/s
- Phi-4
- 39 tokens/s
Time to first token
- GPT-6 Luna
- 95.97 s
- Phi-4
- 2.59 s
DesignArena rating
- GPT-6 Luna
- 1,277
- Phi-4
- Not listed
DesignArena win rate
- GPT-6 Luna
- 48.5%
- Phi-4
- Not listed
Tools & features
| Tools & features | GPT-6 Luna | Phi-4 |
|---|---|---|
| Function calling | Supported | Not listed |
| Structured outputs | Supported | Supported |
| JSON mode | Supported | Supported |
| Reasoning | Supported | Not listed |
| Deterministic seed | Supported | Supported |
| Prompt caching | Supported | Not listed |
Function calling
- GPT-6 Luna
- Supported
- Phi-4
- Not listed
Structured outputs
- GPT-6 Luna
- Supported
- Phi-4
- Supported
JSON mode
- GPT-6 Luna
- Supported
- Phi-4
- Supported
Reasoning
- GPT-6 Luna
- Supported
- Phi-4
- Not listed
Deterministic seed
- GPT-6 Luna
- Supported
- Phi-4
- Supported
Prompt caching
- GPT-6 Luna
- Supported
- Phi-4
- Not listed
API & availability
| API & availability | GPT-6 Luna | Phi-4 |
|---|---|---|
| API identifier | gpt-6-luna | phi-4 |
| API providers | OpenAI, OpenAI, Azure, Azure, Azure, Amazon Bedrock, OpenAI | DeepInfra |
| Knowledge cutoff | Not listed | 2024-06-30 |
| Prices checked | Oct 8, 2026 | Oct 8, 2026 |
API identifier
- GPT-6 Luna
gpt-6-luna- Phi-4
phi-4
API providers
- GPT-6 Luna
- OpenAI, OpenAI, Azure, Azure, Azure, Amazon Bedrock, OpenAI
- Phi-4
- DeepInfra
Knowledge cutoff
- GPT-6 Luna
- Not listed
- Phi-4
- 2024-06-30
Prices checked
- GPT-6 Luna
- Oct 8, 2026
- Phi-4
- Oct 8, 2026
About the models
GPT-6 Luna
Smallest, fastest GPT-6 model for high-volume classification, extraction and chat. Long-context requests bill at $0.20 input / $0.75 output per 1M.
Full pricing & detailsComparison FAQ
GPT-6 Luna: $0.10 input and $0.50 output per million tokens. Phi-4: $0.07 input and $0.14 output per million tokens. The cheaper choice depends on your input-to-output ratio; a model can have a lower input rate but a higher output rate.