Qwen3 235B A22B 2507 Instruct
AlibabaQwenOpen WeightApache 2.0 · Usage Commercial
Description
Qwen3-235B-A22B-Instruct-2507 is the updated instruct version of Qwen3-235B-A22B featuring significant improvements in general capabilities including instruction following, logical reasoning, text comprehension, mathematics, science, coding and tool usage. It provides substantial gains in long-tail knowledge coverage across multiple languages and markedly better alignment with user preferences in subjective and open-ended tasks.
Date de sortie
2025-07-21
Paramètres
235.0B
Longueur du contexte
262K
Modalités
text
Radar de capacités
33
general
52
coding
76
reasoning
56
science
60
agents
0
multimodal
Classements
| Domaine | #Rang | Score | Source |
|---|---|---|---|
| Classement codage | 369 | 39.0 | AA |
| Classement général | 329 | 41.0 | AA |
| Science | 300 | 47.0 | AA |
Scores de benchmarks (LLM Stats)
(LLM Stats (zeroeval))Chat
IFEvalGoogle Research (2023)
88.7%Aut.
Multi-IF
77.5%Aut.
Tau2 Airline
44.0%Aut.
Communication
Tau2 Retail
71.3%Aut.
Creativity
Creative Writing v3
87.5%Aut.
Factuality
SimpleQA
54.3%Aut.
General
MultiPL-E
87.9%Aut.
CSimpleQA
84.3%Aut.
Include
79.5%Aut.
Arena-Hard v2
79.2%Aut.
BFCL-v3
70.9%Aut.
Aider-Polyglot
57.3%Aut.
Language
MMLU-Redux
93.1%Aut.
MMLU-Pro
83.0%Aut.
MMLU-ProX
79.4%Aut.
Math
LiveBench 20241125
75.4%Aut.
AIME 2025
70.3%Aut.
HMMT25
55.4%Aut.
PolyMATH
50.2%Aut.
Reasoning
ZebraLogic
95.0%Aut.
GPQANYU + Cohere + Anthropic (2023)
77.5%Aut.
SuperGPQA
62.6%Aut.
LiveCodeBench v6
51.8%Aut.
ARC-AGI
41.8%Aut.
Writing
WritingBench
85.2%Aut.
Indices d'évaluation AA
(Artificial Analysis)Math 500(OpenAI (2024), subset of Hendrycks et al. MATH (2021))98.0
Mmlu Pro(TIGER-Lab (Univ. of Waterloo, Toronto, CMU, 2024))82.8
Gpqa(NYU + Cohere + Anthropic (2023))75.3
Math Index(Artificial Analysis)71.7
Aime(MAA (Mathematical Association of America))71.7
Aime 25(MAA (Mathematical Association of America))71.7
Livecodebench(UC Berkeley + MIT + Cornell (2024))52.4
Ifbench(Google Research (2023))46.1
Lcr(Artificial Analysis)33.9
Tau2(Sierra + U Toronto + Vector Institute (2025))33.3
Terminalbench Hard(Stanford × Laude Institute (2026))15.2
Intelligence Index(Artificial Analysis)12.0
Hle(Center for AI Safety + Scale AI (2025))11.1
Scores par catégorie LLM Stats
(LLM Stats (zeroeval))Instruction Following80
Language80
Legal80
Structured Output80
Finance80
Healthcare80
Biology80
Creativity80
Writing80
Chat70
Math70
Physics70
Reasoning70
General70
Agents70
Chemistry70
Communication70
Code60
Economics60
Tool Calling60
Multimodal50
Spatial Reasoning50
Factuality50
Vision50
Tarification
Prix d'entrée$0.23 / 1M tokens
Prix de sortie$0.92 / 1M tokens
Prix mixte (3:1)$0.403 / 1M tokens
Prix de lecture cache$0.0175 / 1M tokens
Vitesse
Tokens/sec0.0
Délai du premier token0.00s
Temps de réponse0.00s
Classement des Prix par Fournisseur
Classement des Prix par Fournisseur
11 fournisseurs
Moins cher: DeepInfraPlus cher: GreenPT
FournisseurEntréeSortie
1DeepInfraMoins cher
$0
$0
2Cortecs
$0.069
$0.455
3TensorX
$0.072
$0.464
4OpenRouter
$0.0875
$0.35
5DevPass (LLM Gateway)
$0.09
$0.58
6Kilo Gateway
$0.1495
$0.598
7Venice AI
$0.15
$0.75
8AlibabaPRINCIPAL
$0.23
$0.92
9302.AI
$0.29
$1.143
10Scaleway
$0.75
$2.25
11GreenPT
$1.026
$3.078
Comparer les prix entre différents fournisseurs API pour ce modèle.