MiMo-V2.6-Flash

Runs on a single 8-GPU node (173 GB MXFP4). Xiaomi's own Terminal-Bench 2.1 run is 87.6; the Vals run is shown. Cached input $0.0028.

Released

Sep 21, 2026

Price in / out

$0.14 / $0.28

Blended $/1M

$0.175

Context

1.05M

Parameters: 309B / 15B active. License: MIT.

Benchmark scoresvs MiMo-V2.5-Pro

Benchmark	Score	Δ vs prior	Rank (current)	Configuration	Source
Terminal-Bench 2.1	76.4%	+19.1	#8 of 22	Vals AI run · independent run	vals.ai ↗
Humanity's Last Exam	35.1%	+1.1	#23 of 31	Artificial Analysis run, no tools (2,158 text-only questions) · independent run	artificialanalysis.ai ↗

MiMo versions

Timeline

Version	Released	$ in / out	Terminal-Bench	SWE-bench	GPQA	HLE	τ²-bench	MMLU-Pro
MiMo-V2-FlashSuperseded	2025-12-16	$0.1 / $0.3	—	73.4%	83.7%	—	83.9%	84.9%
MiMo-V2-ProRetired	2026-03-18	$1 / $3	—	78%	87%	—	95%	—
MiMo-V2.5Superseded	2026-04-22	—	60.7%	—	81.6%	—	—	82.9%
MiMo-V2.5-ProSuperseded	2026-04-22	—	57.3%	—	82.6%	34%	94.2%	84.6%
MiMo-V2.6-Flash	2026-09-21	$0.14 / $0.28	76.4%	—	—	35.1%	—	—
MiMo-V2.6-Pro	2026-09-21	$0.435 / $0.87	67.8%	—	—	49.4%	—	—

Compare

Open the Comparison Lab with MiMo-V2.6-Flash and its open-weight peers pre-selected, then toggle any benchmark or model.

Every MiMo version over time

Shipping with it?

Benchmarks are a starting point. If you run MiMo-V2.6-Flash in production, post what it does well and where it falls over; other founders are deciding right now.

Write about MiMo-V2.6-Flash

Current open-weight peers

All

gpt-oss-120bOpenAI Gemma 4 31BGoogle DeepMind Muse Glimmer 30BMeta DeepSeek V4.1 Flash$0.3 in DeepSeek V4 Pro$1.32 in Kimi K3$3 in