Saltar al contenido principal

Qwen3.5 35B A3B (Non-reasoning)

AlibabaQwenOpen WeightApache 2.0 · Uso Comercial

Descripción

Qwen3.5-35B-A3B is a multimodal Mixture-of-Experts model with 35 billion total parameters and 3 billion activated parameters. It combines strong reasoning, coding, agentic, and visual understanding performance with production-friendly efficiency and a native 262K context window.

Fecha de lanzamiento
2026-02-24
Parámetros
35.0B
Longitud del contexto
262K
Modalidades
image, text, video

Radar de capacidades

70
general
60
coding
80
reasoning
68
scienceest.
60
agents
80
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 agéntica83
46.0
LS
Ranking multimodal96
46.0
LS

Puntuaciones de benchmarks (LLM Stats)

(LLM Stats (zeroeval))

Agents

t2-bench81.2%Aut.
VITA-Bench31.9%Aut.
DeepPlanning22.8%Aut.

Chat

IFEvalGoogle Research (2023)91.9%Aut.
Multi-Challenge60.0%Aut.

Code

FullStackBench en58.1%Aut.
FullStackBench zh55.0%Aut.

General

C-Eval90.2%Aut.
MAXIFE86.6%Aut.
Include79.7%Aut.
AndroidWorld_SR71.1%Aut.
NOVA-6357.1%Aut.

Healthcare

PMC-VQA62.0%Aut.
MedXpertQA61.4%Aut.

Instruction Following

IFBench70.2%Aut.

Language

MMLU-Redux93.3%Aut.
MMLU-Pro85.3%Aut.
MMMLU85.2%Aut.
MMLU-ProX81.0%Aut.
WMT24++76.3%Aut.

Long Context

LongBench v259.0%Aut.

Math

HMMT2589.2%Aut.
HMMT 202589.0%Aut.
MathVista-Mini86.2%Aut.
DynaMath85.0%Aut.
MathVision83.9%Aut.
CodeForces0.82 / 3000Aut.
PolyMATH64.4%Aut.

Multimodal

VideoMME w/o sub.82.5%Aut.
MMMU81.4%Aut.
VideoMMMU80.4%Aut.
TIR-Bench55.5%Aut.
OSWorld-Verified54.5%Aut.

Reasoning

Global PIQA86.6%Aut.
GPQANYU + Cohere + Anthropic (2023)84.2%Aut.
CharXiv-R77.5%Aut.
LiveCodeBench v674.6%Aut.
BrowseComp-zh69.5%Aut.
SWE-Bench Verified69.2%Aut.
SuperGPQA63.4%Aut.
BrowseCompOpenAI (2025)61.0%Aut.
AA-LCR58.5%Aut.
Humanity's Last Exam47.4%Aut.
Seal-041.4%Aut.
Terminal-Bench 2.0Stanford × Laude Institute (2026)40.5%Aut.
OJBench36.0%Aut.

Search

WideSearch57.1%Aut.

Tool Calling

BFCL-V467.3%Aut.

Video

MLVU85.6%Aut.

Vision

CountBench0.98 / 100Aut.
VLMsAreBlind97.0%Aut.
V*92.7%Aut.
AI2D92.6%Aut.
MMBench-V1.191.5%Aut.
OCRBench91.0%Aut.
OmniDocBench 1.589.3%Aut.
RefCOCO-avg0.89 / 100Aut.
VideoMME w sub.86.6%Aut.
RealWorldQA84.1%Aut.
EmbSpatialBench0.83 / 100Aut.
MMStar81.9%Aut.
CC-OCR80.7%Aut.
LingoQA79.2%Aut.
SlakeVQA78.7%Aut.
MMMU-Pro75.1%Aut.
MVBench74.8%Aut.
MMVU72.3%Aut.
LVBench71.4%Aut.
ScreenSpot Pro68.6%Aut.
Hallusion Bench67.9%Aut.
ERQA64.8%Aut.
RefSpatialBench0.64 / 100Aut.
MMLongBench-Doc0.59 / 100Aut.
SimpleVQA0.58 / 100Aut.
ODinW42.6%Aut.
BabyVision38.4%Aut.
ZEROBench-Sub0.34 / 100Aut.
SUNRGBD0.33 / 100Aut.
Nuscene14.6%Aut.
Hypersim0.13 / 100Aut.
ZEROBench0.08 / 100Aut.

Índices de evaluación AA

(Artificial Analysis)

No hay datos de evaluación AA disponibles

Puntuaciones por categoría LLM Stats

(LLM Stats (zeroeval))
Chat
80
Image To Text
80
Instruction Following
80
Language
80
Legal
80
Math
80
Physics
80
Structured Output
80
Embodied
80
Finance
80
Biology
80
Text-to-image
80
Video
80
Long Context
70
Multimodal
70
Reasoning
70
Spatial Reasoning
70
Frontend Development
70
General
70
Grounding
70
Healthcare
70
Chemistry
70
Vision
70
Search
60
Code
60
Communication
60
Economics
60
Tool Calling
60
Agents
50
3d
20
Spatial
10

Precios

Precio de entrada$0.25 / 1M tokens
Precio de salida$2 / 1M tokens
Precio mixto (3:1)$0.355 / 1M tokens

Velocidad

Tokens/seg175.4
Retraso del primer token1.99s
Tiempo hasta la respuesta1.99s

Ranking de Precios por Proveedor

Ranking de Precios por Proveedor

6 proveedores

Más barato: AIHubMixMás caro: DevPass (LLM Gateway)
ProveedorEntradaSalida
1AIHubMixMás barato
$0.0564
$0.4512
2302.AI
$0.06
$0.46
3DeepInfra
$0.14
$1
4Requesty
$0.14
$1
5AlibabaPRINCIPAL
$0.25
$2
6DevPass (LLM Gateway)
$0.25
$2

Comparar precios entre diferentes proveedores de API para este modelo.

Fuentes externas