MiMo-V2.5 API Pricing
Xiaomi · released 2026-04-22
Pricing
0.014¢ / 1K
0.028¢ / 1K
0.017¢ / 1K
What it actually costs
| Workload | Tokens | Cost |
|---|---|---|
| Short chat reply | 500 in / 300 out | 0.0154¢ |
| RAG query | 4,000 in / 500 out | 0.07¢ |
| Long document summary | 50,000 in / 1,000 out | 0.728¢ |
| Code review of a 2k-line file | 30,000 in / 2,000 out | 0.476¢ |
| 1,000 chat replies/day for 30 days | 30,000 × (500 in / 300 out) | $4.62 |
Cost = (input tokens ÷ 1,000,000 × $0.14) + (output tokens ÷ 1,000,000 × $0.28).
Benchmarks
#80 of 533
#53 of 197
Similar intelligence, lower price
| Model | Intelligence | Blended $/1M |
|---|---|---|
| Hy3-preview (Reasoning) | 22.7 | $0.10 |
| Ling 3.0 Flash | 24.9 | $0.111 |
| Ling-3.0-flash-VL | 24.6 | $0.111 |
| DeepSeek V4 Flash 0420 (Reasoning, High Effort) | 26.0 | $0.168 |
Leaderboard neighbours
| Model | Intelligence | Blended $/1M |
|---|---|---|
| Hy3 | 25.3 | $0.25 |
| K2 Horizon MoVA 36B A4B | 25.3 | — |
| Grok 4.20 0309 (Reasoning) | 25.2 | $3.00 |
| MiMo-V2.5 | 25.2 | $0.175 |
| Qwen3.7 Plus | 25.2 | $0.70 |
| MiMo-V2-Omni-0327 | 25.1 | — |
| Inkling (xhigh) | 25.0 | $1.762 |
More from Xiaomi
| Model | Intelligence | Blended $/1M |
|---|---|---|
| MiMo-V2.6-Pro | 46.3 | $0.544 |
| MiMo-V2-Pro | 28.6 | — |
| MiMo-V2.5-Pro | 26.0 | $0.544 |
| MiMo-V2-Omni-0327 | 25.1 | — |
| MiMo-V2-Omni | 23.9 | — |
| MiMo-V2-Flash (Feb 2026) | 22.4 | — |
| MiMo-V2-Flash (Reasoning) | 20.8 | $0.15 |
| MiMo-V2-Flash (Non-reasoning) | 16.0 | — |
Count tokens & estimate cost for MiMo-V2.5 See MiMo-V2.5 on the leaderboard
Source: Artificial Analysis · last synced 2026-09-22
FAQ
Who develops MiMo-V2.5?
MiMo-V2.5 is developed by Xiaomi, released 2026-04-22.
How much does MiMo-V2.5 cost per million tokens?
MiMo-V2.5 costs $0.14 per 1M input tokens and $0.28 per 1M output tokens ($0.175 blended), per Artificial Analysis's official pricing data.
How does MiMo-V2.5 rank on benchmarks?
MiMo-V2.5 scores 25.2 on the Artificial Analysis Intelligence Index, ranking #80 of 533 models we track.
Is there a cheaper model with similar intelligence to MiMo-V2.5?
Yes — Hy3-preview (Reasoning) scores similar intelligence (22.7) at $0.10/1M blended, versus $0.175/1M for MiMo-V2.5.
Where does this data come from?
Pricing and benchmark data for MiMo-V2.5 comes from Artificial Analysis (artificialanalysis.ai), synced daily. Last synced 2026-09-22.