Side-by-side comparison of pricing, 12 benchmarks, and generation speed.
| Metric | Qwen3 30B A3B (Non-reasoning) | MiMo-V2.5-Pro |
|---|---|---|
| Input ($/M tokens) | $0.2 | $0.43 |
| Output ($/M tokens) | $0.8 | $0.87 |
Verdict. MiMo-V2.5-Pro takes the aggregate benchmark matchup 2–0 across 2 categories. Real workloads usually care about a handful of specific tasks — see the per-benchmark table above.
Pricing. Both models sit in the budget bracket for output-token pricing. At 0.9× the per-million-token cost, Qwen3 30B A3B (Non-reasoning) is meaningfully cheaper if your traffic is output-heavy (long completions, document generation, agent loops). Qwen3 30B A3B (Non-reasoning) makes more sense when output volume is low and absolute reasoning quality justifies the premium.
Strengths. Qwen3 30B A3B (Non-reasoning) is strongest on Intelligence Index (6.6). MiMo-V2.5-Pro leads on Coding Index (60.2), Intelligence Index (42.9).
Speed. On throughput, Qwen3 30B A3B (Non-reasoning) generates tokens at 107 tok/s versus 59 tok/s — about 45% faster. On time-to-first-token, Qwen3 30B A3B (Non-reasoning) responds in 2200ms vs 3180ms, which matters most for chat-style UIs.
Provider. Alibaba and Xiaomi sell to overlapping but distinct developer audiences: Alibaba tends to ship frontier reasoning models with premium positioning, while Xiaomi often prices more aggressively. Your existing vendor relationships, billing, and SLA preferences may matter as much as the raw numbers above.
Workload cost. Workload scenarios (per million requests at 30M input + 15M output tokens): Qwen3 30B A3B (Non-reasoning) costs $18.00 ($216/year); MiMo-V2.5-Pro costs $25.95 ($311/year). At a smaller 5M-input/2M-output scale (single-developer tool or prototype): Qwen3 30B A3B (Non-reasoning) ≈ $2.60/run, MiMo-V2.5-Pro ≈ $3.89/run. At agent/realtime scale (200M input / 100M output per million requests): Qwen3 30B A3B (Non-reasoning) ≈ $120/run, MiMo-V2.5-Pro ≈ $173/run. Qwen3 30B A3B (Non-reasoning) becomes more attractive at higher volume — the absolute per-token pricing difference compounds when you ship at scale.
Recommendation. Both models have legitimate use cases — the right answer depends on whether you are optimizing for benchmark ceiling, latency, or unit cost. Start with the cheaper / faster model, evaluate against your specific task, and only switch if the upgrade shows a meaningful lift.
Head-to-head deltas
Data from Artificial Analysis API — 12 benchmarks
Qwen3 30B A3B (Non-reasoning) is cheaper overall. Its blended price (3:1 input/output ratio) is $0.35/M tokens vs $0.54/M for MiMo-V2.5-Pro.
MiMo-V2.5-Pro wins 2 out of 12 benchmarks compared to 0 for Qwen3 30B A3B (Non-reasoning). See the detailed benchmark chart above for per-category results.
Qwen3 30B A3B (Non-reasoning) generates tokens faster at 107 tok/s vs 59 tok/s. Qwen3 30B A3B (Non-reasoning) also has lower time-to-first-token (2.20s vs 3.18s).
Choose based on your priorities: Qwen3 30B A3B (Non-reasoning) for lower cost, MiMo-V2.5-Pro for stronger benchmark performance, and Qwen3 30B A3B (Non-reasoning) for faster generation. For latency-sensitive apps, check the TTFT comparison above.