MiMo-V2.5-Pro
AI model by Xiaomi. Real-time pricing and benchmark data.
Benchmarks
Editorial Analysis
Profile. MiMo-V2.5-Pro is a budget-tier option from Xiaomi at $0.87/M output tokens. Despite the low price, it scores competitively on GPQA Diamond (86.6%) and IFBench (79.9%) — making it a candidate for high-volume production workloads where unit cost dominates.
Latency. On latency, MiMo-V2.5-Pro reports 65 tok/s output speed and 2111ms time-to-first-token. Output speed is on the slower end; for real-time chat consider pairing with a streaming front-end or routing latency-sensitive requests to a faster model.
Cost model. For a workload of 1M input tokens + 100K output tokens, MiMo-V2.5-Pro costs $0.52 per million input-equivalent requests. Multiply by your monthly request volume to project the bill, and compare against the speed/correctness profile above to decide if the trade-off is worth it for your workload.
Editorial summary generated from public pricing and benchmark data (Artificial Analysis API). All numbers are sourced — see the Artificial Analysis methodology for benchmark definitions. Last refreshed 2026-07-24.
Compare with similar models
| Model | Input | Output | Speed |
|---|---|---|---|
MiMo-V2.5-ProCurrent | $0.43 | $0.87 | 65 tok/s |
| DeepSeek V4 Pro (Reasoning, High Effort) | $0.43 | $0.87 | 73 tok/s |
| DeepSeek V4 Pro (Reasoning, Max Effort) | $0.43 | $0.87 | 71 tok/s |
| DeepSeek V4 Pro (Non-reasoning) | $0.43 | $0.87 | 73 tok/s |
| Trinity Large Thinking | $0.23 | $0.88 | 169 tok/s |
| Llama 4 Maverick | $0.27 | $0.85 | 99 tok/s |