Qwen2 Instruct 72B
AlibabaQwenOpen Weighttongyi-qianwen
Description
Qwen2-72B-Instruct is an instruction-tuned language model with 72 billion parameters, supporting a context length of up to 131,072 tokens. It's part of the new Qwen2 series, which has surpassed most open-source models and demonstrates competitiveness against proprietary models across various benchmarks.
Date de sortie
2024-06-07
Paramètres
72.0B
Longueur du contexte
—
Modalités
—
Radar de capacités
22
general
17
coding
36
reasoning
25
science
30
agents
0
multimodal
Classements
| Domaine | #Rang | Score | Source |
|---|---|---|---|
| Classement codage | 454 | 17.0 | AA |
| Classement général | 453 | 28.0 | AA |
| Science | 475 | 25.0 | AA |
Scores de benchmarks (LLM Stats)
(LLM Stats (zeroeval))Biology
GPQANYU + Cohere + Anthropic (2023)
42.4%Aut.
Code
HumanEvalOpenAI (2021)
86.0%Aut.
EvalPlus
0.79 / 100Aut.
Finance
MMLU
82.3%Aut.
MMLU-Pro
64.4%Aut.
TruthfulQA
54.8%Aut.
TheoremQA
44.4%Aut.
General
CMMLU
90.1%Aut.
C-Eval
83.8%Aut.
MBPP
0.80 / 100Aut.
MultiPL-E
69.2%Aut.
ARC-C
68.9%Aut.
Language
Winogrande
85.1%Aut.
BBH
82.4%Aut.
Math
GSM8k
91.1%Aut.
MATH
59.7%Aut.
Reasoning
HellaSwagAI2 (2019)
87.6%Aut.
Indices d'évaluation AA
(Artificial Analysis)Intelligence Index(Artificial Analysis)5.7
Math 500(OpenAI (2024), subset of Hendrycks et al. MATH (2021))0.7
Mmlu Pro(TIGER-Lab (Univ. of Waterloo, Toronto, CMU, 2024))0.6
Gpqa(NYU + Cohere + Anthropic (2023))0.4
Scicode(UIUC + Argonne National Lab (2024))0.2
Livecodebench(UC Berkeley + MIT + Cornell (2024))0.2
Aime(MAA (Mathematical Association of America))0.1
Hle(Center for AI Safety + Scale AI (2025))0.0
Scores par catégorie LLM Stats
(LLM Stats (zeroeval))Language80
Code80
Legal70
Math70
Reasoning70
General70
Healthcare70
Finance60
Physics40
Biology40
Chemistry40
Tarification
Prix d'entréeGratuit
Prix de sortieGratuit
Prix mixte (3:1)Gratuit
Vitesse
Tokens/sec0.0
Délai du premier token0.00s
Temps de réponse0.00s
Classement des Prix par Fournisseur
Aucune donnée de fournisseur disponible