Saltar al contenido principal

Kimi K2.6 (Non-reasoning)

KimiKimiOpen WeightModified MIT License

Descripción

Kimi K2.6 is Moonshot AI's open-source, native multimodal agentic model focused on state-of-the-art coding, long-horizon execution, and agent swarm capabilities. It scales horizontally to 300 sub-agents executing 4,000 coordinated steps, dynamically decomposing tasks into parallel, domain-specialized subtasks. K2.6 unifies text, image, and video input with thinking and non-thinking modes, supports a 256K context, and powers proactive 24/7 background agents that manage schedules, execute code, and orchestrate cross-platform operations without human oversight.

Fecha de lanzamiento
2026-04-20
Parámetros
1.0T
Longitud del contexto
262K
Modalidades
image, text, video

Radar de capacidades

32
general
40
coding
79
reasoning
53
science
60
agents
80
multimodal

Rankings

Dominio#PosiciónPuntuaciónFuente
Capacidad agéntica28
56.0
LS
Ranking de codificación130
68.0
AA
Ranking general142
63.0
AA
Razonamiento matemático27
84.0
LB
Ranking multimodal14
63.0
LS
Razonamiento26
79.0
LB
Ciencia168
60.0
AA

Puntuaciones de benchmarks (LLM Stats)

(LLM Stats (zeroeval))

Agents

BrowseCompOpenAI (2025)86.3%Aut.
DeepSearchQA83.0%Aut.
Claw-Eval80.9%Aut.
WideSearch80.8%Aut.
OSWorld-Verified73.1%Aut.
Terminal-Bench 2.0Stanford × Laude Institute (2026)66.7%Aut.
SWE-Bench ProPrinceton NLP (2024)58.6%Aut.
MCP-Mark55.9%Aut.
Toolathlon50.0%Aut.
Finance Agent v244.9%
APEX-Agents27.9%Aut.
FrontierSWE27.0%

Biology

GPQANYU + Cohere + Anthropic (2023)90.5%Aut.
SciCode52.2%Aut.

Code

SWE-Bench Verified80.2%Aut.
SWE-bench Multilingual76.7%Aut.

General

LiveCodeBench v689.6%Aut.
MMMU-Pro80.1%Aut.
LiveBench72.2%

Math

AIME 202696.4%Aut.
MathVision93.2%Aut.
HMMT Feb 2692.7%Aut.
IMO-AnswerBench86.0%Aut.
Humanity's Last Exam36.4%Aut.

Multimodal

V*96.9%Aut.
CharXiv-R86.7%Aut.
BabyVision68.5%Aut.

Reasoning

OJBench60.6%Aut.

Índices de evaluación AA

(Artificial Analysis)
Intelligence Index(Artificial Analysis)
35.4
Tau2(Sierra + U Toronto + Vector Institute (2025))
0.9
Gpqa(NYU + Cohere + Anthropic (2023))
0.8
Lcr(Artificial Analysis)
0.7
Ifbench(Google Research (2023))
0.4
Scicode(UIUC + Argonne National Lab (2024))
0.4
Terminalbench Hard(Stanford × Laude Institute (2026))
0.4
Hle(Center for AI Safety + Scale AI (2025))
0.2

Puntuaciones por categoría LLM Stats

(LLM Stats (zeroeval))
Math
80
Multimodal
80
Search
80
Frontend Development
80
Vision
80
Physics
70
Reasoning
70
General
70
Biology
70
Chemistry
70
Agents
60
Code
60
Tool Calling
60
Finance
40

Precios

Precio de entrada$0.95 / 1M tokens
Precio de salida$4 / 1M tokens
Precio mixto (3:1)$1.712 / 1M tokens
Precio de lectura caché$0.0944 / 1M tokens

Velocidad

Tokens/seg0.0
Retraso del primer token0.00s
Tiempo hasta la respuesta0.00s

Ranking de Precios por Proveedor

Ranking de Precios por Proveedor

17 proveedores

Más barato: DeepInfraMás caro: TensorX
ProveedorEntradaSalida
1DeepInfraMás barato
$0
$0
2Fireworks
$0
$0
3Novita
$0
$0
4Moonshot AI
$0
$0
5Together
$0
$0
6NanoGPT
$0.5
$2.6
7OpenRouter
$0.5605
$2.36
8Lilac
$0.7
$3.5
9FastRouter
$0.75
$3.5
10NovitaAI
$0.8
$3.4
11Kilo Gateway
$0.8
$3.4
12KimiPRINCIPAL
$0.95
$4
13ZenMux
$0.95
$4
14Vercel AI Gateway
$0.95
$4
15Ambient
$0.95
$4
16Ofox
$0.95
$4
17TensorX
$1
$4

Comparar precios entre diferentes proveedores de API para este modelo.

Fuentes externas