Gemini 2.5 Flash-Lite (Reasoning) vs Gemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning)
Comparing 2 AI models · 6 benchmarks · Google
Composite Indices
Intelligence, Coding, Math
Standard Benchmarks
Academic and industry benchmarks
Benchmark Winners
Gemini 2.5 Flash-Lite (Reasoning)
- MATH 500
Gemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning)
- GPQA
- MMLU Pro
- HLE
- LiveCodeBench
- AIME 2025
| Metric | Go Gemini 2.5 Flash-Lite (Reasoning) | Go Gemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning) |
|---|---|---|
| Pricing Per 1M tokens | ||
| Input Cost | $0.10/1M | $0.10/1M |
| Output Cost | $0.40/1M | $0.40/1M |
| Blended Cost 3:1 input/output ratio | $0.18/1M | $0.18/1M |
| Specifications | ||
| Organization Model creator | ||
| Release Date Launch date | Jun 17, 2025 | Sep 8, 2025 |
| Performance & Speed | ||
| Throughput Output speed | — | 5.3 tok/s |
| Time to First Token (TTFT) Initial response delay | — | 3556ms |
| Latency Time to first answer token | — | 3556ms |
| Composite Indices | ||
| Intelligence Index Overall reasoning capability | 40.1 | 47.9 |
| Coding Index Programming ability | 27.6 | 36.5 |
| Math Index Mathematical reasoning | 53.3 | 68.7 |
| Standard Benchmarks | ||
| GPQA Graduate-level reasoning | 62.5% | 70.9% |
| MMLU Pro Advanced knowledge | 75.9% | 80.8% |
| HLE Hard language evaluation | 6.4% | 6.6% |
| LiveCodeBench Real-world coding tasks | 59.3% | 68.8% |
| MATH 500 Mathematical problems | 96.9% | — |
| AIME 2025 Advanced math competition | 53.3% | 68.7% |
| AIME (Original) Math olympiad problems | 70.3% | — |
| SciCode Scientific code generation | 19.3% | 28.7% |
| LCR Code review capability | 51.3% | 59.0% |
| IFBench Instruction-following | 49.9% | 52.6% |
| TAU-bench v2 Tool use & agentic tasks | 18.4% | 30.7% |
| TerminalBench Hard CLI command generation | 4.3% | 12.1% |
Key Takeaways
Gemini 2.5 Flash-Lite (Reasoning) offers the best value at $0.10/1M, making it ideal for high-volume applications and cost-conscious projects.
Gemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning) leads in reasoning capabilities with a 70.9% GPQA score, excelling at complex analytical tasks and problem-solving.
Gemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning) achieves a 36.5 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
Gemini 2.5 Flash-Lite (Reasoning)
- Cost-sensitive applications
- High-volume processing
Gemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning)
- Complex reasoning tasks
- Research & analysis
- Code generation
- Software development
Try Models for Free
Try Gemini 2.5 Flash-Lite (Reasoning) for FREE
No credit card or account required.
Try Gemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning) for FREE
No credit card or account required.
Cost Calculator
Gemini 2.5 Flash-Lite (Reasoning) Gemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning) Costs are estimates based on API pricing. Actual costs may vary based on caching, batch processing, and volume discounts.
Recommended Comparisons
