Model Comparison
Gemini 2.5 Flash-Lite (Non-reasoning)
vs. Step 3.7 Flash
Comparing 2 AI models · 12 benchmarks · Google, StepFun
Recommended Pick
St Step 3.7 Flash 11 metric wins Strongest on: Throughput, Reasoning, Intelligence
Best Value
Gemini 2.5 Flash-Lite (Non-reasoning)
100.0 value score
39.6 reasoning / $0.18/1M
Lowest Price
Gemini 2.5 Flash-Lite (Non-reasoning)
$0.10/1M input price
Best Reasoning
Step 3.7 Flash
47.8 reasoning score
Blends available reasoning benchmarks
Best for Coding
Step 3.7 Flash
37.1 coding index
Composite Indices
Higher is better; speed and price are normalized
Standard Benchmarks
Only benchmarks with data are shown
Differences That Matter
Best value
Gemini 2.5 Flash-Lite (Non-reasoning) has the strongest quality-to-price mix at 100.0 out of 100 value points.
Price gap
Gemini 2.5 Flash-Lite (Non-reasoning) is 2.0x cheaper on input tokens than Step 3.7 Flash.
Speed gap
Step 3.7 Flash generates about 1.7x as many tokens per second as Gemini 2.5 Flash-Lite (Non-reasoning).
Reasoning gap
Step 3.7 Flash leads Gemini 2.5 Flash-Lite (Non-reasoning) by 8.2 points on reasoning.
Coding gap
Step 3.7 Flash leads Gemini 2.5 Flash-Lite (Non-reasoning) by 29.7 points on coding.
Response Face-Off
Run one prompt through the selected models and compare response quality with live speed and cost context.
Gemini 2.5 Flash-Lite (Non-reasoning)
TTFT
—
Time
—
tok/s
—
Tokens
—
Cost
—
Step 3.7 Flash
StepFun
TTFT
—
Time
—
tok/s
—
Tokens
—
Cost
—
Which answer was more useful?
Full Comparison
| Metric | Go Gemini 2.5 Flash-Lite (Non-reasoning) | Top Pick St Step 3.7 Flash |
|---|---|---|
| Pricing per 1M tokens | ||
| Input Cost | $0.10/1M | $0.20/1M |
| Output Cost | $0.40/1M | $1.15/1M |
| Blended (3:1) | $0.18/1M | $0.44/1M |
| Specifications | ||
| Organization | StepFun | |
| Release Date | Jun 17, 2025 | May 29, 2026 |
| Performance & Speed | ||
| Throughput | 235.5 tok/s | 397.1 tok/s |
| TTFT | 359ms | 718ms |
| Latency | 359ms | 5755ms |
| Composite Indices | ||
| Value Score | 100.0 | 48.3 |
| Reasoning Score | 39.6 | 47.8 |
| Intelligence | 12.7 | 42.6 |
| Coding | 7.4 | 37.1 |
| Math | 35.3 | — |
| Standard Benchmarks | ||
| GPQA | 47.4% | 80.9% |
| MMLU Pro | 72.4% | — |
| HLE | 3.7% | 19.9% |
| LiveCodeBench | 40.0% | — |
| MATH 500 | 92.6% | — |
| AIME 2025 | 35.3% | — |
| AIME (Original) | 50.0% | — |
| SciCode | 17.7% | 40.0% |
| LCR | 31.3% | 63.7% |
| IFBench | 31.5% | 67.3% |
| TAU-bench v2 | 19.0% | 98.5% |
| TerminalBench Hard | 2.3% | 35.6% |
Key Takeaways
Gemini 2.5 Flash-Lite (Non-reasoning) offers the best value at $0.10/1M, making it ideal for high-volume applications and cost-conscious projects.
Step 3.7 Flash has the strongest reasoning profile with a 47.8 reasoning score, combining the available reasoning-heavy benchmarks.
Step 3.7 Flash reaches a 37.1 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
Gemini 2.5 Flash-Lite (Non-reasoning)
- Cost-sensitive applications
- High-volume processing
Step 3.7 Flash
- Complex reasoning tasks
- Research & analysis
- Code generation
- Software development
