Model Comparison
Small 3
vs. Step 3.7 Flash
Comparing 2 AI models · 12 benchmarks · Mistral, StepFun
Recommended Pick
St Step 3.7 Flash 9 metric wins Strongest on: Throughput, Reasoning, Intelligence
Best Value
Small 3
100.0 value score
21.6 reasoning / $0.10/1M
Lowest Price
Small 3
$0.07/1M 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
Small 3 has the strongest quality-to-price mix at 100.0 out of 100 value points.
Price gap
Small 3 is 2.7x cheaper on input tokens than Step 3.7 Flash.
Speed gap
Step 3.7 Flash generates about 2.5x as many tokens per second as Small 3.
Reasoning gap
Step 3.7 Flash leads Small 3 by 26.2 points on reasoning.
Top-pick rationale
Step 3.7 Flash wins 9 measurable categories, including Throughput, Reasoning, Intelligence, GPQA.
Response Face-Off
Run one prompt through the selected models and compare response quality with live speed and cost context.
Small 3
Mistral
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 | Mi Small 3 | Top Pick St Step 3.7 Flash |
|---|---|---|
| Pricing per 1M tokens | ||
| Input Cost | $0.07/1M | $0.20/1M |
| Output Cost | $0.19/1M | $1.15/1M |
| Blended (3:1) | $0.10/1M | $0.44/1M |
| Specifications | ||
| Organization | Mistral | StepFun |
| Release Date | Jan 30, 2025 | May 29, 2026 |
| Performance & Speed | ||
| Throughput | 158.9 tok/s | 397.1 tok/s |
| TTFT | 445ms | 721ms |
| Latency | 445ms | 5758ms |
| Composite Indices | ||
| Value Score | 100.0 | 52.5 |
| Reasoning Score | 21.6 | 47.8 |
| Intelligence | 12.7 | 42.6 |
| Coding | — | 37.1 |
| Math | 4.3 | — |
| Standard Benchmarks | ||
| GPQA | 46.2% | 80.9% |
| MMLU Pro | 65.2% | — |
| HLE | 4.1% | 19.9% |
| LiveCodeBench | 25.2% | — |
| MATH 500 | 71.5% | — |
| AIME 2025 | 4.3% | — |
| AIME (Original) | 8.0% | — |
| SciCode | 23.6% | 40.0% |
| LCR | 0.0% | 63.7% |
| IFBench | 26.4% | 67.3% |
| TAU-bench v2 | 19.6% | 98.5% |
| TerminalBench Hard | — | 35.6% |
Key Takeaways
Small 3 offers the best value at $0.07/1M, 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
Small 3
- Cost-sensitive applications
- High-volume processing
Step 3.7 Flash
- Complex reasoning tasks
- Research & analysis
- Code generation
- Software development
