MiMo-V2.6-Flash

MiMo-V2.6-Flash is the efficiency-oriented member of Xiaomi's MiMo-V2.6 series, released and open-sourced on September 22, 2026 alongside the more capable MiMo-V2.6-Pro and the speed-optimized MiMo-V2.6-Pro-UltraSpeed. Xiaomi positions the series as natively omnimodal and frames MiMo-V2.6-Flash specifically as the variant that strikes the best balance between intelligence, efficiency, and cost, making it the practical choice for users who want frontier-style capabilities without the heavier compute footprint of the Pro tier.

The MiMo-V2.6 series reflects Xiaomi's continued exploration of what the team calls the RSI path, scaling reinforcement learning compute on verifiable, complex tasks so the model can keep expanding its capability frontier through exploration and feedback. On Xiaomi's own release-page scorecards, MiMo-V2.6-Flash posts a DeepSWE v1.1 score of 67.9, 26.0 on ProgramBench, and 61.2 on the in-house MiMo Code Bench, with third-party LLM Stats rankings placing it competitively in coding (top 10%) and in the middle of the pack for tool calling, reasoning, and vision. That profile points to a model well suited for coding assistants, agentic tool use, and general omnimodal workloads where a strong intelligence-to-cost ratio matters more than peak accuracy.

OpenCode Gomimo-v2.6-flashmimo

Open provider page Provider docs

Quick Info

Provider: OpenCode Go
Model key: mimo-v2.6-flash
Release date: Sep 22, 2026
Last updated: Sep 22, 2026
Input modalities
Output modalities
Capabilities

Cost

Input token cost: $0.14
Output token cost: $0.28

Limits

Output tokens: 131,072 tokens
Context window: 1,048,576 tokens

Transparent token rates

Compare MiMo-V2.6-Flash pricing

Rates are shown per one million tokens. Combined means one million input plus one million output tokens.

Browse this family

Latest news about MiMo-V2.6-Flash

OpenCode Go

CoverageBenchmark

The eesel AI blog confirms Xiaomi released the MiMo V2.6 family — including the Flash variant — as open-weight, natively omnimodal language models on September 22, 2026. Every checkpoint in the family, Flash included, ships with a 1-million-token context window and handles text, images, video, and audio natively rather The eesel piece provides the family-level architecture summary from the official MiMo-V2.6-Pro model card on Hugging Face: a hybrid sliding-window/global attention backbone on a sparse MoE, with GA/SWA transformer blocks and a multi-token-prediction decoder. While detailed parameter counts and benchmark numbers in the

OpenCode Go

Coverage

Xiaomi released MiMo-V2.6-Flash alongside its flagship MiMo-V2.6-Pro on September 22, 2026, positioning Flash as a smaller, cheaper sibling targeted at high-volume production workloads. Per VentureBeat's reporting, Flash retains the same 1-million-token context window and native multimodal capabilities (text, image, au VentureBeat details Xiaomi's API pricing for MiMo-V2.6-Flash at $0.14 per million uncached input tokens and $0.28 per million output tokens — substantially below the Pro tier at $0.435/$0.87. While the headline Artificial Analysis Intelligence Index score of 46 and the ~134 tokens/second output speed figure are attribu