MiMo-V2.5 API Pricing
MiMo-V2.5 was Xiaomi's omni-modal budget tier from April 2026, and it is on its way out. Xiaomi deprecated it with the V2.6 release on September 22, 2026 and shuts the id down at 10:00 Beijing time on October 21, with no automatic routing: requests that still name it will fail. Until then it bills $0.14/M input, $0.0028/M cached input and $0.28/M output, exactly what MiMo-V2.6-Flash costs, so switching changes the model name and nothing on the invoice.
Run the numbers.
Calculator pre-loaded with MiMo-V2.5's USD rates, which stay in force until the id shuts down on October 21, 2026. The same figures apply to MiMo-V2.6-Flash, which also has a Batch API at half.
Real-world presets.
Repo repair task
Analyst memo draft
Web research brief
Product support turn
This is a chars-per-token approximation, not a real tokenizer. Actual tokens vary by language, code density, and tool-call overhead — counts are typically ±10–20% off for English prose, more for code or non-Latin scripts. For exact billing, use the vendor's official tokenizer.
| Model | Input /M | Output /M | Effective blended | Context | Best for |
|---|---|---|---|---|---|
| MiMo-V2.5 Current | $0.14 cache $0.0028 | $0.28 | $0.0477 agentic 92/8 | 1M | Budget omni-modal / text-capable agent loops |
| MiMo-V2.6-Flash | $0.14 cache $0.0028 | $0.28 | $0.0477 same | 1M | The successor, at the same price |
| MiMo-V2.6-Pro | $0.435 cache $0.0036 | $0.87 | $0.144 +3.0× | 1M | Xiaomi's flagship, for harder agent runs |
| MiniMax M3 | $0.30 cache $0.06 | $1.20 | $0.191 +4.0× | 1M | MiniMax frontier coding agents |
| DeepSeek V4 Pro | $1.32 cache $0.044 | $3.96 | $0.569 +12× | 1M | Low-cost reasoning workloads |
| Qwen3 Max | $1.20 | $6.00 | $1.58 deprecated · +33× | 252K | Alibaba flagship workloads |
Frequently asked.
Practical questions about MiMo-V2.5 pricing, cache hits, context size, and cheap 1M-context MiMo workloads.
Q · 01 Is MiMo-V2.5 being retired? +
mimo-v2.5 down at 10:00 Beijing time on October 21, 2026 (02:00 UTC) and lists no system replacement, so from then on requests that name it return an error. In June the retired V2 ids were routed to V2.5 automatically; this time they are not. The successor is MiMo-V2.6-Flash, which the pricing table prints on the same row at the same rates.Q · 02 What is MiMo-V2.5 priced at? +
MiMo-V2.5 at $0.14/M input, $0.0028/M cached input, and $0.28/M output.Q · 03 How is the effective price calculated? +
92% input, 8% output, and 82% input cache hits. For MiMo-V2.5, that lands at about $0.05/M blended tokens.Q · 04 What context window does MiMo-V2.5 support? +
1M context for the V2.5 family row used on this page.Q · 05 Does this page model media/audio/image pricing? +
Q · 06 When was this price last checked? +
2026-09-24, against Xiaomi's pay-as-you-go pricing page and its model-deprecation page. The rates have not changed since the model was added to this catalogue in April 2026.