Model Comparison
Ministral 3 14B
vs. Medium 3
Comparing 2 AI models · 12 benchmarks · Mistral
Recommended Pick
Mi Medium 3 11 metric wins Strongest on: Reasoning, Intelligence, Coding
Best Value
Ministral 3 14B
100.0 value score
27.6 reasoning / $0.20/1M
Lowest Price
Ministral 3 14B
$0.20/1M input price
Best Reasoning
Medium 3
39.5 reasoning score
Blends available reasoning benchmarks
Best for Coding
Medium 3
13.6 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
Ministral 3 14B has the strongest quality-to-price mix at 100.0 out of 100 value points.
Price gap
Ministral 3 14B is 2.0x cheaper on input tokens than Medium 3.
Speed gap
Ministral 3 14B generates about 2.1x as many tokens per second as Medium 3.
Reasoning gap
Medium 3 leads Ministral 3 14B by 11.9 points on reasoning.
Coding gap
Medium 3 leads Ministral 3 14B by 2.7 points on coding.
Response Face-Off
Run one prompt through the selected models and compare response quality with live speed and cost context.
Ministral 3 14B
Mistral
TTFT
—
Time
—
tok/s
—
Tokens
—
Cost
—
Medium 3
Mistral
TTFT
—
Time
—
tok/s
—
Tokens
—
Cost
—
Which answer was more useful?
Full Comparison
| Metric | Mi Ministral 3 14B | Top Pick Mi Medium 3 |
|---|---|---|
| Pricing per 1M tokens | ||
| Input Cost | $0.20/1M | $0.40/1M |
| Output Cost | $0.20/1M | $2.00/1M |
| Blended (3:1) | $0.20/1M | $0.80/1M |
| Specifications | ||
| Organization | Mistral | Mistral |
| Release Date | Dec 2, 2025 | May 7, 2025 |
| Performance & Speed | ||
| Throughput | 95.1 tok/s | 45.8 tok/s |
| TTFT | 426ms | 560ms |
| Latency | 426ms | 560ms |
| Composite Indices | ||
| Value Score | 100.0 | 35.8 |
| Reasoning Score | 27.6 | 39.5 |
| Intelligence | 16.0 | 18.8 |
| Coding | 10.9 | 13.6 |
| Math | 30.0 | 30.3 |
| Standard Benchmarks | ||
| GPQA | 57.2% | 57.8% |
| MMLU Pro | 69.3% | 76.0% |
| HLE | 4.6% | 4.3% |
| LiveCodeBench | 35.1% | 40.0% |
| MATH 500 | — | 90.7% |
| AIME 2025 | 30.0% | 30.3% |
| AIME (Original) | — | 44.0% |
| SciCode | 23.6% | 33.1% |
| LCR | 22.0% | 28.0% |
| IFBench | 32.0% | 39.3% |
| TAU-bench v2 | 27.2% | 24.3% |
| TerminalBench Hard | 4.5% | 3.8% |
Key Takeaways
Ministral 3 14B offers the best value at $0.20/1M, making it ideal for high-volume applications and cost-conscious projects.
Medium 3 has the strongest reasoning profile with a 39.5 reasoning score, combining the available reasoning-heavy benchmarks.
Medium 3 reaches a 13.6 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
Ministral 3 14B
- Cost-sensitive applications
- High-volume processing
Medium 3
- Complex reasoning tasks
- Research & analysis
- Code generation
- Software development
