MiMo API Providers
Who serves MiMo, at what price and speed.
MiMo is Xiaomi's open-weight model family (V2.5 and V2.6, Flash and Pro). Xiaomi serves most MiMo tokens itself.
Xiaomi served 91% of the 13.1T MiMo tokens that went through a large public LLM router over Oct 3 to Oct 9, 2026. Xiaomi's own API served 91%; 11 providers list the family.
Who serves MiMo tokens
| Provider | Tokens, 7d | Share |
|---|---|---|
| Xiaomi (lab) | 11.8T | 90.8% |
| NovitaAI | 638B | 4.9% |
| GMICloud | 297B | 2.3% |
| DeepInfra | 81B | 0.6% |
| io.net | 55B | 0.4% |
| inference.net | 47B | 0.4% |
| Makora | 47B | 0.4% |
| Venice | 29B | 0.2% |
| DigitalOcean | 8B | <0.1% |
| Darkbloom | 955M | <0.1% |
| Relace | 0 | 0.0% |
| AtlasCloud | 0 | 0.0% |
| StreamLake | 0 | 0.0% |
All MiMo models combined, Oct 3 to Oct 9, 2026.
MiMo-V2.6-Flash: prices and speed by provider
Output prices run from $0.28 to $0.41 per million tokens across 8 providers, median $0.28.
| Provider | Share | Input $/M | Output $/M | Cache read $/M | Tokens/s | Latency | Uptime 3d | Context |
|---|---|---|---|---|---|---|---|---|
| Xiaomi | 94.8% | $0.14 | $0.28 | $0.0028 | 36 | 3.25s | 99.9% | 1.05M |
| NovitaAI | 3.0% | $0.14 | $0.28 | $0.0028 | 30 | 2.69s | 96.7% | 1.05M |
| io.net | 0.5% | $0.26 | $0.41 | $0.025 | 47 | 0.82s | 99.7% | 1.05M |
| GMICloud | 0.5% | $0.14 | $0.28 | $0.0030 | 21 | 3.90s | 79.7% | 1.05M |
| Makora | 0.4% | $0.13 | $0.28 | $0.0020 | 69 | 0.38s | 99.0% | 1.05M |
| DeepInfra | 0.3% | $0.14 | $0.28 | $0.0028 | 25 | 0.94s | 99.6% | 1.05M |
| Venice | 0.1% | $0.175 | $0.35 | $0.0038 | 16 | 1.46s | 98.5% | 1M |
| Darkbloom | <0.1% | $0.10 | $0.28 | $0.05 | 17 | 2.59s | 98.6% | 1.05M |
List prices as of October 9, 2026, cheapest endpoint per provider. Tokens/s and latency are medians over 30 minutes.
MiMo-V2.6-Pro: prices and speed by provider
Output prices run from $0.87 to $0.87 per million tokens across 4 providers, median $0.87.
| Provider | Share | Input $/M | Output $/M | Cache read $/M | Tokens/s | Latency | Uptime 3d | Context |
|---|---|---|---|---|---|---|---|---|
| Xiaomi | 52.9% | $0.435 | $0.87 | $0.0036 | 31 | 3.33s | 96.7% | 1.05M |
| NovitaAI | 25.7% | $0.435 | $0.87 | $0.0036 | 34 | 2.96s | 94.9% | 1.05M |
| GMICloud | 17.3% | $0.435 | $0.87 | $0.0040 | 40 | 4.42s | 88.9% | 1.05M |
| DeepInfra | 4.1% | $0.43 | $0.87 | $0.0036 | 44 | 1.51s | 100.0% | 1.05M |
List prices as of October 9, 2026, cheapest endpoint per provider. Tokens/s and latency are medians over 30 minutes.
All MiMo models
| Model | Tokens, 7d | Providers | Output $/M | Context | Released |
|---|---|---|---|---|---|
| MiMo-V2.6-Flash | 10.9T | 8 | $0.28 to $0.41 | 1.05M | 2026-09-21 |
| MiMo-V2.6-Pro | 1.24T | 4 | $0.87 | 1.05M | 2026-09-21 |
| MiMo-V2.5 | 773B | 5 | $0.238 to $2.00 | 1.05M | 2026-04-22 |
| MiMo-V2.5-Pro | 174B | 6 | $0.609 to $1.80 | 1.05M | 2026-04-22 |
Finance your GPUs
Serving MiMo on your own GPUs? Tell us the cluster, the models you serve and the contracts behind them, and we will share it with partner funders that fit.
Prefer email? hello@amcompute.com