Passer au contenu principal
AITier

Qwen3.5 397B A17B (Non-reasoning)

AlibabaQwenOpen WeightApache 2.0 · Usage Commercial

Description

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.

Date de sortie
2026-02-16
Paramètres
397.0B
Longueur du contexte
262K
Modalités
audio, image, text, video

Radar de capacités

30
general
41
coding
86
reasoning
57
science
60
agents
70
multimodal

Classements

Domaine#RangScoreSource
Capacité agentique29
56.0
LS
Classement codage142
64.0
AA
Classement général151
61.0
AA
Science126
64.0
AA

Scores de benchmarks (LLM Stats)

(LLM Stats (zeroeval))

Agents

t2-bench86.7%Aut.
WideSearch74.0%Aut.
BFCL-V472.9%Aut.
BrowseCompOpenAI (2025)69.0%Aut.
Terminal-Bench 2.0Stanford × Laude Institute (2026)52.5%Aut.
VITA-Bench49.7%Aut.
MCP-Mark46.1%Aut.
Toolathlon38.3%Aut.
DeepPlanning34.3%Aut.

Biology

GPQANYU + Cohere + Anthropic (2023)88.4%Aut.

Chemistry

SuperGPQA70.4%Aut.

Code

SWE-Bench Verified76.4%Aut.
SWE-bench Multilingual69.3%Aut.
SecCodeBench68.3%Aut.

Communication

Multi-Challenge67.6%Aut.

Finance

MMLU-Pro87.8%Aut.
MMLU-ProX84.7%Aut.

General

MMLU-Redux94.9%Aut.
C-Eval93.0%Aut.
IFEvalGoogle Research (2023)92.6%Aut.
Global PIQA89.8%Aut.
MMMLU88.5%Aut.
MAXIFE88.2%Aut.
Include85.6%Aut.
LiveCodeBench v683.6%Aut.
IFBench76.5%Aut.
LongBench v263.2%Aut.
NOVA-6359.1%Aut.

Language

WMT24++78.9%Aut.

Long Context

AA-LCR68.7%Aut.

Math

HMMT 202594.8%Aut.
HMMT2592.7%Aut.
AIME 202691.3%Aut.
IMO-AnswerBench80.9%Aut.
PolyMATH73.3%Aut.
Humanity's Last Exam28.7%Aut.

Reasoning

BrowseComp-zh70.3%Aut.
Seal-046.9%Aut.

Indices d'évaluation AA

(Artificial Analysis)
Intelligence Index(Artificial Analysis)
32.7
Gpqa(NYU + Cohere + Anthropic (2023))
0.9
Tau2(Sierra + U Toronto + Vector Institute (2025))
0.8
Lcr(Artificial Analysis)
0.6
Ifbench(Google Research (2023))
0.5
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

Scores par catégorie LLM Stats

(LLM Stats (zeroeval))
Language
90
Biology
90
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

Tarification

Prix d'entrée$0.6 / 1M tokens
Prix de sortie$3.6 / 1M tokens
Prix mixte (3:1)$1.35 / 1M tokens

Vitesse

Tokens/sec78.6
Délai du premier token1.36s
Temps de réponse1.36s

Classement des Prix par Fournisseur

Classement des Prix par Fournisseur

1 fournisseurs

FournisseurEntréeSortie
1AlibabaPRINCIPAL
$0.6
$3.6

Comparer les prix entre différents fournisseurs API pour ce modèle.

Sources externes

LLM StatsArtificial Analysis