Model Comparison
MiMo-V2.5-Pro
vs. Qwen3 8B (Non-reasoning)
Comparing 2 AI models · 12 benchmarks · Xiaomi, Alibaba
Recommended Pick
Xi MiMo-V2.5-Pro 10 metric wins Strongest on: Reasoning, Intelligence, Coding
Best Value
Qwen3 8B (Non-reasoning)
100.0 value score
29.8 reasoning / $0.18/1M
Lowest Price
Qwen3 8B (Non-reasoning)
$0.18/1M input price
Best Reasoning
MiMo-V2.5-Pro
54.2 reasoning score
Blends available reasoning benchmarks
Best for Coding
MiMo-V2.5-Pro
45.5 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
Qwen3 8B (Non-reasoning) has the strongest quality-to-price mix at 100.0 out of 100 value points.
Price gap
Qwen3 8B (Non-reasoning) is 2.4x cheaper on input tokens than MiMo-V2.5-Pro.
Speed gap
Qwen3 8B (Non-reasoning) generates about 1.3x as many tokens per second as MiMo-V2.5-Pro.
Reasoning gap
MiMo-V2.5-Pro leads Qwen3 8B (Non-reasoning) by 24.4 points on reasoning.
Coding gap
MiMo-V2.5-Pro leads Qwen3 8B (Non-reasoning) by 38.4 points on coding.
Response Face-Off
Run one prompt through the selected models and compare response quality with live speed and cost context.
MiMo-V2.5-Pro
Xiaomi
TTFT
—
Time
—
tok/s
—
Tokens
—
Cost
—
Qwen3 8B (Non-reasoning)
Alibaba
TTFT
—
Time
—
tok/s
—
Tokens
—
Cost
—
Which answer was more useful?
Full Comparison
| Metric | Top Pick Xi MiMo-V2.5-Pro | Al Qwen3 8B (Non-reasoning) |
|---|---|---|
| Pricing per 1M tokens | ||
| Input Cost | $0.43/1M | $0.18/1M |
| Output Cost | $0.87/1M | $0.20/1M |
| Blended (3:1) | $0.54/1M | $0.18/1M |
| Specifications | ||
| Organization | Xiaomi | Alibaba |
| Release Date | Apr 22, 2026 | Apr 28, 2025 |
| Performance & Speed | ||
| Throughput | 47.5 tok/s | 63.6 tok/s |
| TTFT | 1832ms | 1407ms |
| Latency | 43974ms | 1407ms |
| Composite Indices | ||
| Value Score | 61.8 | 100.0 |
| Reasoning Score | 54.2 | 29.8 |
| Intelligence | 42.2 | 5.1 |
| Coding | 45.5 | 7.1 |
| Math | — | 24.3 |
| Standard Benchmarks | ||
| GPQA | 86.6% | 45.2% |
| MMLU Pro | — | 64.3% |
| HLE | 33.8% | 2.8% |
| LiveCodeBench | — | 20.2% |
| MATH 500 | — | 82.8% |
| AIME 2025 | — | 24.3% |
| AIME (Original) | — | 24.3% |
| SciCode | 50.2% | 16.8% |
| LCR | 73.3% | 0.0% |
| IFBench | 79.9% | 28.6% |
| TAU-bench v2 | 94.2% | 24.9% |
| TerminalBench Hard | 43.2% | 2.3% |
Key Takeaways
Qwen3 8B (Non-reasoning) offers the best value at $0.18/1M, making it ideal for high-volume applications and cost-conscious projects.
MiMo-V2.5-Pro has the strongest reasoning profile with a 54.2 reasoning score, combining the available reasoning-heavy benchmarks.
MiMo-V2.5-Pro reaches a 45.5 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
- Complex reasoning tasks
- Research & analysis
- Code generation
- Software development
Qwen3 8B (Non-reasoning)
- Cost-sensitive applications
- High-volume processing
