Gemini 2.5 Flash Preview (Sep '25) (Non-reasoning) vs Grok 4
Comparing 2 AI models · 6 benchmarks · Google, xAI
Most Affordable
Go
Gemini 2.5 Flash Preview (Sep '25) (Non-reasoning)
$0.30/1M
Highest Intelligence
xA
Grok 4
87.7% GPQA
Best for Coding
xA
Grok 4
55.1 Coding Index
Price Difference
10.0x
input cost range
Composite Indices
Intelligence, Coding, Math
Standard Benchmarks
Academic and industry benchmarks
Benchmark Winners
6 tests
Go
Gemini 2.5 Flash Preview (Sep '25) (Non-reasoning)
0
No clear wins
xA
Grok 4
6
- GPQA
- MMLU Pro
- HLE
- LiveCodeBench
- MATH 500
- AIME 2025
| Metric | Go Gemini 2.5 Flash Preview (Sep '25) (Non-reasoning) | xA Grok 4 |
|---|---|---|
| Pricing Per 1M tokens | ||
| Input Cost | $0.30/1M | $3.00/1M |
| Output Cost | $2.50/1M | $15.00/1M |
| Blended Cost 3:1 input/output ratio | $0.85/1M | $6.00/1M |
| Specifications | ||
| Organization Model creator | xAI | |
| Release Date Launch date | Sep 25, 2025 | Jul 10, 2025 |
| Performance & Speed | ||
| Throughput Output speed | 237.7 tok/s | 37.2 tok/s |
| Time to First Token (TTFT) Initial response delay | 345ms | 9172ms |
| Latency Time to first answer token | 345ms | 9172ms |
| Composite Indices | ||
| Intelligence Index Overall reasoning capability | 46.7 | 65.3 |
| Coding Index Programming ability | 37.8 | 55.1 |
| Math Index Mathematical reasoning | 56.7 | 92.7 |
| Standard Benchmarks | ||
| GPQA Graduate-level reasoning | 76.6% | 87.7% |
| MMLU Pro Advanced knowledge | 83.6% | 86.6% |
| HLE Hard language evaluation | 7.8% | 23.9% |
| LiveCodeBench Real-world coding tasks | 62.5% | 81.9% |
| MATH 500 Mathematical problems | — | 99.0% |
| AIME 2025 Advanced math competition | 56.7% | 92.7% |
| AIME (Original) Math olympiad problems | — | 94.3% |
| SciCode Scientific code generation | 37.5% | 45.7% |
| LCR Code review capability | 56.7% | 68.0% |
| IFBench Instruction-following | 43.5% | 53.7% |
| TAU-bench v2 Tool use & agentic tasks | 28.4% | 74.9% |
| TerminalBench Hard CLI command generation | 13.5% | 37.6% |
Key Takeaways
Gemini 2.5 Flash Preview (Sep '25) (Non-reasoning) offers the best value at $0.30/1M, making it ideal for high-volume applications and cost-conscious projects.
Grok 4 leads in reasoning capabilities with a 87.7% GPQA score, excelling at complex analytical tasks and problem-solving.
Grok 4 achieves a 55.1 coding index, making it the top choice for software development and code generation tasks.
All models support context windows of ∞+ tokens, suitable for processing lengthy documents and maintaining extended conversations.
When to Choose Each Model
Go