MiMo-V2.6-Pro-UltraSpeed is Xiaomi's optimized inference variant of the MiMo-V2.6-Pro foundation model, tuned for faster throughput while maintaining output quality.
- Strengths
- Delivers roughly 10x faster inference speed compared to the standard MiMo-V2.6-Pro while preserving the same model quality and capabilities.
- Best for
- Workloads requiring low-latency responses from a high-capability model, or high-throughput scenarios where speed is prioritized alongside quality.
- Limitations
- As an optimized variant rather than a fundamental architectural change, it sits between the earlier MiMo-V2-Flash tier and the full MiMo-V2.6-Pro in the lineup; use the standard version if inference speed is not a constraint.
Input / 1M
$4.35
Output / 1M
$8.70
Cached input / 1M
$0.036
Context window
1.05M
Price history
Snapshots
| Effective | Input | Output | Cached in | Note | Source |
|---|---|---|---|---|---|
| 22 Sep 2026 | $4.35 | $8.70 | $0.036 | Imported from OpenRouter | openrouter.ai |