Saltar al contenido principal

MiMo-V2.6-Pro

XiaomiOpen WeightMIT · Uso Comercial

Descripción

MiMo-V2.6-Pro is Xiaomi's MIT-licensed omnimodal sparse Mixture-of-Experts model with 1.02T total parameters and 42B 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-Pro-RL. Benchmark results are Xiaomi-reported release evaluations, distinct from intermediate live-training checkpoints. UltraSpeed serves the same model at higher speed for $4.35 input and $8.70 output per million tokens.

Fecha de lanzamiento
2026-09-21
Parámetros
1.0T
Longitud del contexto
1.0M
Modalidades
audio, image, text, video

Radar de capacidades

47
general
61
coding
60
reasoning
57
science
60
agents
80
multimodal

Rankings

Dominio#PosiciónPuntuaciónFuente
Capacidad agéntica38
51.0
LS
Ranking de codificación4
97.0
AA
Ranking general24
79.0
AA
Ranking multimodal41
59.0
LS
Ciencia28
85.0
AA

Puntuaciones de benchmarks (LLM Stats)

(LLM Stats (zeroeval))

Agents

Toolathlon-Verified76.9%Aut.
SEC-bench Pro66.3%Aut.
JobBench62.0%Aut.
AutomationBench v1.0.653.1%Aut.
Agents' Last Exam31.6%Aut.
Program Bench26.5%Aut.
ExploitGym17.8%Aut.

Code

CyberGym94.0%Aut.
MiMo Cyber Bench81.7%Aut.
MiMo Visual Coding72.3%Aut.
DeepSWE 1.171.9%Aut.
MiMo Coding Bench63.2%Aut.
ExploitBench47.9%Aut.

General

GDPval-AA 2.11673.00 / 3000Aut.
AA Intelligence Index v4.346.3%Aut.

Multimodal

OSWorld-Verified82.0%Aut.

Reasoning

Terminal-Bench 2.189.9%Aut.
Terminal-Bench 4.034.9%Aut.

Índices de evaluación AA

(Artificial Analysis)
Lcr(Artificial Analysis)
86.3
Scicode(UIUC + Argonne National Lab (2024))
60.9
Hle(Center for AI Safety + Scale AI (2025))
49.4
Intelligence Index(Artificial Analysis)
46.3

Puntuaciones por categoría LLM Stats

(LLM Stats (zeroeval))
Multimodal
80
Vision
80
Reasoning
60
Safety
60
General
60
Agents
60
Code
60
Tool Calling
60

Precios

Precio de entrada$0.435 / 1M tokens
Precio de salida$0.87 / 1M tokens
Precio mixto (3:1)$0.544 / 1M tokens
Precio de lectura caché$0.0036 / 1M tokens

Velocidad

Tokens/seg40.2
Retraso del primer token2.76s
Tiempo hasta la respuesta52.55s

Ranking de Precios por Proveedor

Ranking de Precios por Proveedor

14 proveedores

Más barato: XiaomiMás caro: above.dev
ProveedorEntradaSalida
1XiaomiMás barato
$0
$0
2Requesty
$0.43
$0.87
3NanoGPT
$0.435
$0.87
4OpenRouter
$0.435
$0.87
5OpenCode Go
$0.435
$0.87
6ZenMux
$0.435
$0.87
7Kilo Gateway
$0.435
$0.87
8Vercel AI Gateway
$0.435
$0.87
9DevPass (LLM Gateway)
$0.435
$0.87
10EmpirioLabs AI
$0.435
$0.87
11LLM Gateway
$0.435
$0.87
12CrossModel
$0.47
$0.94
13AIHubMix
$0.48
$0.96
14above.dev
$0.5077
$1.0154

Comparar precios entre diferentes proveedores de API para este modelo.

Fuentes externas