Model Comparison
MiMo-V2.5-Pro (Non-reasoning)
vs. Medium 3
Comparing 2 AI models · 12 benchmarks · Xiaomi, Mistral
Recommended Pick
Xi MiMo-V2.5-Pro (Non-reasoning) 10 metric wins Strongest on: Reasoning, Intelligence, Coding
Best Value
Medium 3
100.0 value score
39.5 reasoning / $0.80/1M
Lowest Price
Medium 3
$0.40/1M input price
Best Reasoning
MiMo-V2.5-Pro (Non-reasoning)
41.7 reasoning score
Blends available reasoning benchmarks
Best for Coding
MiMo-V2.5-Pro (Non-reasoning)
36.8 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
Medium 3 has the strongest quality-to-price mix at 100.0 out of 100 value points.
Price gap
Medium 3 is 2.3x cheaper on input tokens than MiMo-V2.5-Pro (Non-reasoning).
Reasoning gap
MiMo-V2.5-Pro (Non-reasoning) leads Medium 3 by 2.2 points on reasoning.
Coding gap
MiMo-V2.5-Pro (Non-reasoning) leads Medium 3 by 23.2 points on coding.
Top-pick rationale
MiMo-V2.5-Pro (Non-reasoning) wins 10 measurable categories, including Reasoning, Intelligence, Coding, GPQA.
Response Face-Off
Run one prompt through the selected models and compare response quality with live speed and cost context.
MiMo-V2.5-Pro (Non-reasoning)
Xiaomi
TTFT
—
Time
—
tok/s
—
Tokens
—
Cost
—
Medium 3
Mistral
TTFT
—
Time
—
tok/s
—
Tokens
—
Cost
—
Which answer was more useful?
Full Comparison
| Metric | Top Pick Xi MiMo-V2.5-Pro (Non-reasoning) | Mi Medium 3 |
|---|---|---|
| Pricing per 1M tokens | ||
| Input Cost | $0.90/1M | $0.40/1M |
| Output Cost | $2.70/1M | $2.00/1M |
| Blended (3:1) | $1.35/1M | $0.80/1M |
| Specifications | ||
| Organization | Xiaomi | Mistral |
| Release Date | Apr 22, 2026 | May 7, 2025 |
| Performance & Speed | ||
| Throughput | 45.7 tok/s | 46.9 tok/s |
| TTFT | 2094ms | 521ms |
| Latency | 2094ms | 521ms |
| Composite Indices | ||
| Value Score | 62.6 | 100.0 |
| Reasoning Score | 41.7 | 39.5 |
| Intelligence | 35.6 | 18.8 |
| Coding | 36.8 | 13.6 |
| Math | — | 30.3 |
| Standard Benchmarks | ||
| GPQA | 76.2% | 57.8% |
| MMLU Pro | — | 76.0% |
| HLE | 13.3% | 4.3% |
| LiveCodeBench | — | 40.0% |
| MATH 500 | — | 90.7% |
| AIME 2025 | — | 30.3% |
| AIME (Original) | — | 44.0% |
| SciCode | 39.1% | 33.1% |
| LCR | 35.0% | 28.0% |
| IFBench | 42.7% | 39.3% |
| TAU-bench v2 | 72.5% | 24.3% |
| TerminalBench Hard | 35.6% | 3.8% |
Key Takeaways
Medium 3 offers the best value at $0.40/1M, making it ideal for high-volume applications and cost-conscious projects.
MiMo-V2.5-Pro (Non-reasoning) has the strongest reasoning profile with a 41.7 reasoning score, combining the available reasoning-heavy benchmarks.
MiMo-V2.5-Pro (Non-reasoning) reaches a 36.8 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
MiMo-V2.5-Pro (Non-reasoning)
- Complex reasoning tasks
- Research & analysis
- Code generation
- Software development
Medium 3
- Cost-sensitive applications
- High-volume processing
