Model Comparison
Mercury 2
vs. Step 3.5 Flash
Comparing 2 AI models · 7 benchmarks · Inception, StepFun
Recommended Pick
St Step 3.5 Flash 14 metric wins Strongest on: Value, Input price, Output price
Best Value
Step 3.5 Flash
100.0 value score
46.7 reasoning / $0.15/1M
Lowest Price
Step 3.5 Flash
$0.10/1M input price
Best Reasoning
Step 3.5 Flash
46.7 reasoning score
Blends available reasoning benchmarks
Best for Coding
Step 3.5 Flash
31.6 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
Step 3.5 Flash has the strongest quality-to-price mix at 100.0 out of 100 value points.
Price gap
Step 3.5 Flash is 2.5x cheaper on input tokens than Mercury 2.
Speed gap
Mercury 2 generates about 5.3x as many tokens per second as Step 3.5 Flash.
Reasoning gap
Step 3.5 Flash leads Mercury 2 by 4.9 points on reasoning.
Coding gap
Step 3.5 Flash leads Mercury 2 by 1.0 points on coding.
Response Face-Off
Run one prompt through the selected models and compare response quality with live speed and cost context.
Mercury 2
Inception
TTFT
—
Time
—
tok/s
—
Tokens
—
Cost
—
Step 3.5 Flash
StepFun
TTFT
—
Time
—
tok/s
—
Tokens
—
Cost
—
Which answer was more useful?
Full Comparison
| Metric | In Mercury 2 | Top Pick St Step 3.5 Flash |
|---|---|---|
| Pricing per 1M tokens | ||
| Input Cost | $0.25/1M | $0.10/1M |
| Output Cost | $0.75/1M | $0.30/1M |
| Blended (3:1) | $0.38/1M | $0.15/1M |
| Specifications | ||
| Organization | Inception | StepFun |
| Release Date | Feb 20, 2026 | Feb 2, 2026 |
| Performance & Speed | ||
| Throughput | 968.4 tok/s | 182.4 tok/s |
| TTFT | 3507ms | 1044ms |
| Latency | 3507ms | 12006ms |
| Composite Indices | ||
| Value Score | 35.8 | 100.0 |
| Reasoning Score | 41.8 | 46.7 |
| Intelligence | 32.8 | 37.8 |
| Coding | 30.6 | 31.6 |
| Standard Benchmarks | ||
| GPQA | 77.0% | 83.1% |
| HLE | 15.5% | 19.1% |
| SciCode | 38.7% | 40.4% |
| LCR | 36.3% | 43.0% |
| IFBench | 69.8% | 64.6% |
| TAU-bench v2 | 70.8% | 94.4% |
| TerminalBench Hard | 26.5% | 27.3% |
Key Takeaways
Step 3.5 Flash offers the best value at $0.10/1M, making it ideal for high-volume applications and cost-conscious projects.
Step 3.5 Flash has the strongest reasoning profile with a 46.7 reasoning score, combining the available reasoning-heavy benchmarks.
Step 3.5 Flash reaches a 31.6 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
Mercury 2
- General-purpose AI
- Versatile applications
Step 3.5 Flash
- Cost-sensitive applications
- High-volume processing
- Complex reasoning tasks
- Research & analysis
- Code generation
- Software development
