MiMo-V2.5 pricing & benchmarks
Released Apr 22, 2026 · served by 7 providers · ranked #39 by intelligence in our index
What MiMo-V2.5 is
MiMo-V2.5 is a native omnimodal model by Xiaomi. It delivers Pro-level agentic performance at roughly half the inference cost, while surpassing MiMo-V2-Omni in multimodal perception across image and video understanding...
- Model ID:
xiaomi/mimo-v2.5 - Modality: text+image+audio+video->text
- Tool calling: supported · Structured output: supported
- Weights:
XiaomiMiMo/MiMo-V2.5on Hugging Face
Providers and prices for MiMo-V2.5
The same model costs different amounts depending on who serves it. Prices are per 1M tokens, cheapest first. Uptime and throughput are OpenRouter's rolling measurements, not vendor claims.
| Provider | Input | Output | Context | Quant | Uptime 24h | Throughput |
|---|---|---|---|---|---|---|
| GMICloudcheapest | $0.112 | $0.224 | 1.1M | fp8 | 90.1% | — |
| Xiaomi | $0.140 | $0.280 | 1.0M | fp8 | 99.0% | — |
| Parasail | $0.140 | $0.280 | 1.0M | fp8 | 98.0% | — |
| Venice | $0.140 | $0.280 | 1M | fp8 | 97.0% | — |
| Novita | $0.168 | $0.336 | 1.0M | fp8 | 98.7% | — |
| Io Net | $0.168 | $0.336 | 262K | fp8 | 97.7% | — |
| DeepInfra | $0.400 | $2.00 | 262K | bf16 | 98.4% | — |
MiMo-V2.5 benchmark results
Design Arena head-to-head results, as reported through the OpenRouter model API.
| Category | Arena | Rank | Elo | Win rate |
|---|---|---|---|---|
| dataviz | models | #16 | 1284 | 55.8% |
| codecategories | models | #17 | 1292 | 55.4% |
| gamedev | models | #18 | 1299 | 56.5% |
| website | models | #18 | 1295 | 55.8% |
| uicomponent | models | #20 | 1295 | 55.9% |
| svg | models | #22 | 1221 | 52.8% |
| 3d | models | #26 | 1274 | 52.7% |
| asciiart | models | #37 | 1172 | 45.9% |
Cheaper models in the same class
Models scoring within 4 points of MiMo-V2.5 on the Intelligence Index, but with a lower output price.
FAQ
How much does the MiMo-V2.5 API cost?
$0.112 per 1M input tokens and $0.224 per 1M output tokens. A typical workload of 1M input + 300K output tokens costs about $0.18.
Is MiMo-V2.5 free?
No. It is a paid model starting at $0.112 per 1M input tokens, though some providers offer trial credits.
Which provider is cheapest for MiMo-V2.5?
GMICloud at $0.112 input / $0.224 output per 1M tokens (fp8 quantization).
Can I self-host MiMo-V2.5?
Yes — weights are published as XiaomiMiMo/MiMo-V2.5 on Hugging Face, so you can run it on your own hardware.