Saltar al contenido principal
AITier

Qwen3.5 397B A17B (Non-reasoning)

AlibabaQwenOpen WeightApache 2.0 · Uso Comercial

Descripción

Qwen3.5-397B-A17B is Qwen's flagship Mixture-of-Experts model with 397 billion total parameters and 17 billion activated parameters. It delivers state-of-the-art performance across knowledge, reasoning, coding, mathematics, multilingual understanding, instruction following, long context, and agent tasks.

Fecha de lanzamiento
2026-02-16
Parámetros
397.0B
Longitud del contexto
262K
Modalidades
audio, image, text, video

Radar de capacidades

21
general
70
coding
86
reasoning
66
science
60
agents
70
multimodal

Rankings

Dominio#PosiciónPuntuaciónFuente
Capacidad agéntica17
62.0
LS
Ranking de codificación216
64.0
AA
Ranking general211
52.0
AA
Ciencia195
60.0
AA

Puntuaciones de benchmarks (LLM Stats)

(LLM Stats (zeroeval))

Agents

t2-bench86.7%Aut.
VITA-Bench49.7%Aut.
MCP-Mark46.1%Aut.
Toolathlon38.3%Aut.
DeepPlanning34.3%Aut.

Chat

IFEvalGoogle Research (2023)92.6%Aut.
Multi-Challenge67.6%Aut.

Code

SecCodeBench68.3%Aut.

General

C-Eval93.0%Aut.
MAXIFE88.2%Aut.
Include85.6%Aut.
NOVA-6359.1%Aut.

Instruction Following

IFBench76.5%Aut.

Language

MMLU-Redux94.9%Aut.
MMMLU88.5%Aut.
MMLU-Pro87.8%Aut.
MMLU-ProX84.7%Aut.
WMT24++78.9%Aut.

Long Context

LongBench v263.2%Aut.

Math

HMMT 202594.8%Aut.
HMMT2592.7%Aut.
AIME 202691.3%Aut.
IMO-AnswerBench80.9%Aut.
PolyMATH73.3%Aut.

Reasoning

Global PIQA89.8%Aut.
GPQANYU + Cohere + Anthropic (2023)88.4%Aut.
LiveCodeBench v683.6%Aut.
SWE-Bench Verified76.4%Aut.
SuperGPQA70.4%Aut.
BrowseComp-zh70.3%Aut.
SWE-bench Multilingual69.3%Aut.
BrowseCompOpenAI (2025)69.0%Aut.
AA-LCR68.7%Aut.
Terminal-Bench 2.0Stanford × Laude Institute (2026)52.5%Aut.
Seal-046.9%Aut.
Humanity's Last Exam28.7%Aut.

Search

WideSearch74.0%Aut.

Tool Calling

BFCL-V472.9%Aut.

Índices de evaluación AA

(Artificial Analysis)
Gpqa(NYU + Cohere + Anthropic (2023))
86.1
Tau2(Sierra + U Toronto + Vector Institute (2025))
83.9
Lcr(Artificial Analysis)
64.3
Ifbench(Google Research (2023))
51.6
Terminalbench Hard(Stanford × Laude Institute (2026))
35.6
Intelligence Index(Artificial Analysis)
21.4
Hle(Center for AI Safety + Scale AI (2025))
19.8

Puntuaciones por categoría LLM Stats

(LLM Stats (zeroeval))
Language
90
Biology
90
Chat
80
Instruction Following
80
Legal
80
Math
80
Physics
80
Structured Output
80
Finance
80
Frontend Development
80
Healthcare
80
Chemistry
80
Long Context
70
Multimodal
70
Reasoning
70
Search
70
Spatial Reasoning
70
General
70
Code
70
Communication
70
Economics
70
Agents
60
Tool Calling
60
Vision
50

Precios

Precio de entrada$0.6 / 1M tokens
Precio de salida$3.6 / 1M tokens
Precio mixto (3:1)$1.35 / 1M tokens

Velocidad

Tokens/seg86.1
Retraso del primer token1.70s
Tiempo hasta la respuesta1.70s

Ranking de Precios por Proveedor

Ranking de Precios por Proveedor

2 proveedores

Más barato: DeepInfraMás caro: Alibaba
ProveedorEntradaSalida
1DeepInfraMás barato
$0
$0
2AlibabaPRINCIPAL
$0.6
$3.6

Comparar precios entre diferentes proveedores de API para este modelo.

Fuentes externas

LLM StatsArtificial Analysis