---
id: "xiaomi/mimo-v2.6-flash"
also: ["xiaomi/mimo-v2.6-flash-20260921", "XiaomiMiMo/MiMo-V2.6-Flash-RL", "hf/XiaomiMiMo/MiMo-V2.6-Flash-RL", "requesty/xiaomi/mimo-v2.6-flash", "vercel/xiaomi/mimo-v2.6-flash"]
card_usd_per_m: "0.14 in, 0.28 out, at xiaomi/fp8"
cheapest_list_usd_per_m: "0.1 in, 0.28 out, at darkbloom/fp4 on openrouter"
cheapest_effective_usd_per_m: "0.011 in, 0.2233 out, at gmicloud/bf16 on openrouter"
context: 1050000
max_output: 131072
input: ["text", "image", "video", "audio"]
output: ["text"]
released: "2026-09-21"
api: ["/api/v1/chat/completions", "/api/v1/completions", "/api/v1/messages", "/api/v1/responses"]
params: ["frequency_penalty", "include_reasoning", "logit_bias", "logprobs", "max_tokens", "min_p", "parallel_tool_calls", "presence_penalty", "reasoning", "repetition_penalty", "response_format", "seed", "stop", "structured_outputs", "temperature", "tool_choice", "tools", "top_k", "top_logprobs", "top_p"]
endpoints: 20
gateways: "openrouter (9), vercel (7), requesty (3), hf (1)"
no_training: 12
trains: 5
retains_prompts: 11
intelligence: 26.7000
---
# Xiaomi: MiMo-V2.6-Flash

MiMo-V2.6-Flash is the efficiency-balanced model in Xiaomi's MiMo-V2.6 series: a sparse Mixture-of-Experts model with 309B total and 15B activated parameters, natively omnimodal across text, image, video and audio, with a 1M-token context window. Trained with large-scale mixed reinforcement learning across coding, general agents, visual and cybersecurity tasks, it is built for agentic workloads such as long-horizon coding, tool use and computer use.