Saltar al contenido principal
AITier

Qwen3.8 Flash

Alibaba Cloud / Qwen TeamQwenProprietary

Descripción

Qwen3.8 Flash is the production QwenCloud / OpenRouter API model (id qwen3.8-flash), not the open-weight Qwen3.8-Flash-Next checkpoint. Hugging Face states Flash is the official managed version based on Flash-Next with production features such as 1M context by default and official built-in tools. Architecture inherits Flash-Next (125B parameters with 6B activated, plus 51B n-gram embedding and 4B MTP). Multimodal text, image, and video understanding; thinking on by default; function calling, built-in tools, and structured output. QwenCloud lists 1M context, ~128k max output, and ~256k thinking budget. Open weights remain under the separate catalog card qwen3.8-flash-next.

Fecha de lanzamiento
2026-08-26
Parámetros
125.0B
Longitud del contexto
1.0M
Modalidades
image, text, video

Radar de capacidades

70
general
60
coding
70
reasoning
77
scienceest.
60
agents
70
multimodal

Science se estima a partir de las puntuaciones científicas de LLM Stats o del razonamiento cuando no hay benchmarks científicos dedicados.

Rankings

Dominio#PosiciónPuntuaciónFuente
Capacidad agéntica4
67.0
LS
Ranking multimodal33
60.0
LS

Puntuaciones de benchmarks (LLM Stats)

(LLM Stats (zeroeval))

Agents

AndroidWorld84.5%Aut.
CoWorkBench73.9%Aut.
Toolathlon73.5%Aut.
ClawEval-MM60.4%Aut.
Job Bench55.7%Aut.
Agents' Last Exam51.2%Aut.
RecreationBench49.9%Aut.

Code

Vision2Web64.0%Aut.
DeepSWE 1.158.7%Aut.
NL2Repo48.1%Aut.

Instruction Following

IFBench81.3%Aut.

Math

MathVision95.7%Aut.

Multimodal

OSWorld 2.019.4%Aut.

Reasoning

LiveCodeBench v691.9%Aut.
GPQANYU + Cohere + Anthropic (2023)91.7%Aut.
CharXiv-R90.6%Aut.
SWE-bench Multilingual81.0%Aut.
SWE-Bench ProPrinceton NLP (2024)62.5%Aut.
Humanity's Last Exam35.9%Aut.

Vision

RealWorldQA88.5%Aut.
LVBench76.6%Aut.
ERQA72.3%Aut.

Índices de evaluación AA

(Artificial Analysis)

No hay datos de evaluación AA disponibles

Puntuaciones por categoría LLM Stats

(LLM Stats (zeroeval))
Physics
90
Biology
90
Chemistry
90
Long Context
80
Spatial Reasoning
80
Instruction Following
80
Math
70
Multimodal
70
Reasoning
70
General
70
Vision
70
Productivity
60
Agents
60
Code
60
Tool Calling
60

Precios

Precio de entrada$0 / 1M tokens
Precio de salida$0 / 1M tokens
Precio mixto (3:1)$0 / 1M tokens
Precio de lectura caché$0.016 / 1M tokens
Precio de escritura caché$0.2 / 1M tokens

Velocidad

No hay datos de velocidad disponibles

Ranking de Precios por Proveedor

Ranking de Precios por Proveedor

12 proveedores

Más barato: Alibaba Cloud / Qwen TeamMás caro: Eden AI
ProveedorEntradaSalida
1Alibaba Cloud / Qwen TeamPRINCIPAL
$0
$0
2Novita
$0
$0
3Alibaba (China)
$0.11875
$0.40073
4Vancine
$0.12
$0.38
5CrossModel
$0.13
$0.43
6Alibaba
$0.15
$0.47
7OpenRouter
$0.15
$0.47
8OpenCode Go
$0.15
$0.47
9Kilo Gateway
$0.15
$0.47
10DevPass (LLM Gateway)
$0.15
$0.47
11Charm Hyper
$0.15
$0.47
12Eden AI
$0.16
$0.47

Comparar precios entre diferentes proveedores de API para este modelo.

Fuentes externas

LLM StatsArtificial Analysis