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
23
general
60
coding
79
reasoning
61
science
60
agents
80
multimodal
Rankings
| Dominio | #Posición | Puntuación | Fuente |
|---|---|---|---|
| Capacidad agéntica | 35 | 52.0 | LS |
| Ranking de codificación | 196 | 68.0 | AA |
| Ranking general | 198 | 54.0 | AA |
| Razonamiento matemático | 46 | 84.0 | LB |
| Ranking multimodal | 51 | 57.0 | LS |
| Razonamiento | 45 | 79.0 | LB |
| Ciencia | 223 | 56.0 | AA |
Puntuaciones de benchmarks (LLM Stats)
(LLM Stats (zeroeval))Agents
MCP-Mark
55.9%Aut.
Toolathlon
50.0%Aut.
Finance Agent v2
44.9%
Code
Claw-Eval
80.9%Aut.
FrontierSWE
27.0%
Math
AIME 2026
96.4%Aut.
MathVision
93.2%Aut.
HMMT Feb 26
92.7%Aut.
IMO-AnswerBench
86.0%Aut.
LiveBench
72.2%
Multimodal
OSWorld-Verified
73.1%Aut.
Reasoning
GPQANYU + Cohere + Anthropic (2023)
90.5%Aut.
LiveCodeBench v6
89.6%Aut.
CharXiv-R
86.7%Aut.
BrowseCompOpenAI (2025)
86.3%Aut.
SWE-Bench Verified
80.2%Aut.
SWE-bench Multilingual
76.7%Aut.
Terminal-Bench 2.0Stanford × Laude Institute (2026)
66.7%Aut.
OJBench
60.6%Aut.
SWE-Bench ProPrinceton NLP (2024)
58.6%Aut.
SciCode
52.2%Aut.
Humanity's Last Exam
36.4%Aut.
APEX-Agents
27.9%Aut.
Search
DeepSearchQA
83.0%Aut.
WideSearch
80.8%Aut.
Vision
V*
96.9%Aut.
MMMU-Pro
80.1%Aut.
BabyVision
68.5%Aut.
Índices de evaluación AA
(Artificial Analysis)Tau2(Sierra + U Toronto + Vector Institute (2025))93.9
Gpqa(NYU + Cohere + Anthropic (2023))78.8
Lcr(Artificial Analysis)69.7
Ifbench(Google Research (2023))44.3
Terminalbench Hard(Stanford × Laude Institute (2026))37.9
Intelligence Index(Artificial Analysis)23.6
Hle(Center for AI Safety + Scale AI (2025))19.6
Puntuaciones por categoría LLM Stats
(LLM Stats (zeroeval))Math80
Multimodal80
Search80
Frontend Development80
Vision80
Physics70
Reasoning70
General70
Biology70
Chemistry70
Agents60
Code60
Tool Calling60
Finance40
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$0.15 / 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
2Novita
$0
$0
3Fireworks
$0
$0
4Moonshot AI
$0
$0
5NanoGPT
$0.5
$2.6
6OpenRouter
$0.65
$3.41
7Kilo Gateway
$0.65
$3.41
8Lilac
$0.7
$3.5
9FastRouter
$0.75
$3.5
10NovitaAI
$0.8
$3.4
11KimiPRINCIPAL
$0.95
$4
12ZenMux
$0.95
$4
13Vercel AI Gateway
$0.95
$4
14Ambient
$0.95
$4
15Ofox
$0.95
$4
16TokenGo
$0.95
$4
17TensorX
$1
$4
Comparar precios entre diferentes proveedores de API para este modelo.