Models, measured side by side
Grok 4.7 vs Phi-4
Compare Grok 4.7, Phi-4 by token pricing, context length, benchmark results, speed and tool support.
Monthly workload
Monthly cost breakdown
InputOutput
Grok 4.7
$70 / mo
Phi-4
$2.10 / mo
Performance comparison
Grok 4.7
46.4
Phi-4
5.9
Overview
Pricing & capacity
| Pricing & capacity | Grok 4.7 | Phi-4 |
|---|---|---|
| Input / 1M tokens | $2.00 | $0.07 |
| Output / 1M tokens | $6.00 | $0.14 |
| Cached input / 1M | $0.50 | Not listed |
| Monthly cost | $70 | $2.10 |
| Context window | 500K | 16K |
| Maximum output | 450K | 15K |
Input / 1M tokens
- Grok 4.7
- $2.00
- Phi-4
- $0.07
Output / 1M tokens
- Grok 4.7
- $6.00
- Phi-4
- $0.14
Cached input / 1M
- Grok 4.7
- $0.50
- Phi-4
- Not listed
Monthly cost
- Grok 4.7
- $70
- Phi-4
- $2.10
Context window
- Grok 4.7
- 500K
- Phi-4
- 16K
Maximum output
- Grok 4.7
- 450K
- Phi-4
- 15K
Benchmarks & performance
| Benchmarks & performance | Grok 4.7 | Phi-4 |
|---|---|---|
| Arena Elo | Not listed | 1,123 |
| MMLU-Pro | Not listed | 60.7% |
| Intelligence Index | 46.4 | 5.9 |
| GPQA | Not listed | 57.5% |
| Humanity’s Last Exam | 43.1% | 3.8% |
| SciCode | 57.4% | Not listed |
| Output speed | 83 tokens/s | 39 tokens/s |
| Time to first token | 43.17 s | 2.59 s |
| DesignArena rating | 1,269 | Not listed |
| DesignArena win rate | 47.6% | Not listed |
Arena Elo
- Grok 4.7
- Not listed
- Phi-4
- 1,123
MMLU-Pro
- Grok 4.7
- Not listed
- Phi-4
- 60.7%
Intelligence Index
- Grok 4.7
- 46.4
- Phi-4
- 5.9
GPQA
- Grok 4.7
- Not listed
- Phi-4
- 57.5%
Humanity’s Last Exam
- Grok 4.7
- 43.1%
- Phi-4
- 3.8%
SciCode
- Grok 4.7
- 57.4%
- Phi-4
- Not listed
Output speed
- Grok 4.7
- 83 tokens/s
- Phi-4
- 39 tokens/s
Time to first token
- Grok 4.7
- 43.17 s
- Phi-4
- 2.59 s
DesignArena rating
- Grok 4.7
- 1,269
- Phi-4
- Not listed
DesignArena win rate
- Grok 4.7
- 47.6%
- Phi-4
- Not listed
Tools & features
| Tools & features | Grok 4.7 | Phi-4 |
|---|---|---|
| Function calling | Supported | Not listed |
| Structured outputs | Supported | Supported |
| JSON mode | Supported | Supported |
| Reasoning | Supported | Not listed |
| Log probabilities | Supported | Not listed |
| Deterministic seed | Supported | Supported |
| Prompt caching | Supported | Not listed |
Function calling
- Grok 4.7
- Supported
- Phi-4
- Not listed
Structured outputs
- Grok 4.7
- Supported
- Phi-4
- Supported
JSON mode
- Grok 4.7
- Supported
- Phi-4
- Supported
Reasoning
- Grok 4.7
- Supported
- Phi-4
- Not listed
Log probabilities
- Grok 4.7
- Supported
- Phi-4
- Not listed
Deterministic seed
- Grok 4.7
- Supported
- Phi-4
- Supported
Prompt caching
- Grok 4.7
- Supported
- Phi-4
- Not listed
API & availability
| API & availability | Grok 4.7 | Phi-4 |
|---|---|---|
| API identifier | grok-4.7 | phi-4 |
| API providers | xAI, xAI, xAI, xAI, xAI | DeepInfra |
| Knowledge cutoff | Not listed | 2024-06-30 |
| Prices checked | Oct 8, 2026 | Oct 8, 2026 |
API identifier
- Grok 4.7
grok-4.7- Phi-4
phi-4
API providers
- Grok 4.7
- xAI, xAI, xAI, xAI, xAI
- Phi-4
- DeepInfra
Knowledge cutoff
- Grok 4.7
- Not listed
- Phi-4
- 2024-06-30
Prices checked
- Grok 4.7
- Oct 8, 2026
- Phi-4
- Oct 8, 2026
About the models
Grok 4.7
Grok 4.7 is SpaceXAI's flagship model for coding, agentic tasks, and knowledge work, succeeding Grok 4.6. It is particularly strong at long-running software engineering tasks, verifying its own work, and...
Full pricing & detailsComparison FAQ
Grok 4.7: $2.00 input and $6.00 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.