Standard & Paid Price
| Meter | Standard Price | Paid Price | Save |
|---|---|---|---|
| Input price | $0.1400 / 1M tokens | $0.1260 / 1M tokens | -10% |
| Completion price | $0.2800 / 1M tokens | $0.2520 / 1M tokens | -10% |
| Cache read price | $0.0028 / 1M tokens | $0.00252 / 1M tokens | -10% |
Description
Xiaomi MiMo-V2.6-Flash — efficient, low-cost reasoning model with open weights, natively full-modality: one model understands images, video, audio and text. 1M context and 128K max output, with deep thinking, function calling and structured output. Aimed at high-volume, everyday professional workloads where cost and throughput matter.
Capabilities
- Tasks
- chat, reasoning, vision, file-video-understanding, audio-understanding
Decision evidence
MiMo-V2.6-Flash is Xiaomi's MIT-licensed open-weight mixture-of-experts model (309B total and 15B active per the model card) with a 1M-token context window and text, image, video and audio input. Reviewed results are split: one-prompt builds rate its output highly at very low cost, but reviewers report long thinking times and stalls on longer agentic work, including an engineer's git task that went badly off course.
2026-09-25 · Markdown evidence · JSON data
API endpoints
POST /v1/chat/completions(openai)POST /v1/messages(anthropic)POST /v1/responses(openai-response)