MiMo-V2.6-Flash
XiaomiOpen WeightMIT · Usage Commercial
Description
MiMo-V2.6-Flash is Xiaomi's MIT-licensed omnimodal sparse Mixture-of-Experts model with 309B total parameters and 15B active parameters. It supports text, image, video, and audio understanding with a 1,048,576-token context window and text output. MiMo-V2.6 scales reinforcement learning across coding, general agents, visual tasks, and cybersecurity. Xiaomi's launch article is dated September 22, 2026; released weights are named MiMo-V2.6-Flash-RL. Benchmark results are Xiaomi-reported release evaluations, distinct from intermediate live-training checkpoints.
Date de sortie
2026-09-21
Paramètres
309.0B
Longueur du contexte
1.0M
Modalités
audio, image, text, video
Radar de capacités
37
general
51
coding
50
reasoning
45
science
50
agents
80
multimodal
Classements
| Domaine | #Rang | Score | Source |
|---|---|---|---|
| Capacité agentique | 105 | 28.0 | LS |
| Classement codage | 98 | 84.0 | AA |
| Classement général | 119 | 64.0 | AA |
| Classement multimodal | 57 | 56.0 | LS |
| Science | 159 | 65.0 | AA |
Scores de benchmarks (LLM Stats)
(LLM Stats (zeroeval))Agents
Toolathlon-Verified
73.6%Aut.
JobBench
61.2%Aut.
AutomationBench v1.0.6
52.3%Aut.
SEC-bench Pro
47.5%Aut.
Agents' Last Exam
27.6%Aut.
Program Bench
26.0%Aut.
ExploitGym
6.0%Aut.
Code
CyberGym
95.1%Aut.
MiMo Cyber Bench
77.2%Aut.
MiMo Visual Coding
71.5%Aut.
DeepSWE 1.1
67.9%Aut.
MiMo Coding Bench
61.2%Aut.
ExploitBench
25.3%Aut.
Multimodal
OSWorld-Verified
80.8%Aut.
Reasoning
Terminal-Bench 2.1
87.6%Aut.
Terminal-Bench 4.0
28.8%Aut.
Indices d'évaluation AA
(Artificial Analysis)Lcr(Artificial Analysis)74.3
Scicode(UIUC + Argonne National Lab (2024))51.3
Intelligence Index(Artificial Analysis)37.9
Hle(Center for AI Safety + Scale AI (2025))35.1
Scores par catégorie LLM Stats
(LLM Stats (zeroeval))Multimodal80
Vision80
Reasoning50
General50
Agents50
Code50
Tool Calling50
Safety40
Tarification
Prix d'entrée$0.14 / 1M tokens
Prix de sortie$0.28 / 1M tokens
Prix mixte (3:1)$0.175 / 1M tokens
Prix de lecture cache$0.0028 / 1M tokens
Vitesse
Tokens/sec47.5
Délai du premier token4.07s
Temps de réponse46.20s
Classement des Prix par Fournisseur
Classement des Prix par Fournisseur
13 fournisseurs
Moins cher: XiaomiPlus cher: above.dev
FournisseurEntréeSortie
1XiaomiMoins cher
$0
$0
2Kilo Gateway
$0.08
$1.28
3NanoGPT
$0.14
$0.28
4OpenRouter
$0.14
$0.28
5OpenCode Go
$0.14
$0.28
6Requesty
$0.14
$0.28
7Vercel AI Gateway
$0.14
$0.28
8DevPass (LLM Gateway)
$0.14
$0.28
9EmpirioLabs AI
$0.14
$0.28
10LLM Gateway
$0.14
$0.28
11AIHubMix
$0.155
$0.31
12CrossModel
$0.16
$0.32
13above.dev
$0.1692
$0.3385
Comparer les prix entre différents fournisseurs API pour ce modèle.