Passer au contenu principal
AITier

Llama 3.1 Instruct 405B

MetaLlamaOpen WeightLlama 3.1 Community License

Description

Llama 3.1 405B Instruct is a large language model optimized for multilingual dialogue use cases. It outperforms many available open source and closed chat models on common industry benchmarks. The model supports 8 languages and has a 128K token context length.

Date de sortie
2024-07-23
Paramètres
405.0B
Longueur du contexte
—
Modalités
text

Radar de capacités

27
general
31
coding
23
reasoning
37
science
70
agents
0
multimodal

Classements

Domaine#RangScoreSource
Classement codage477
24.0
AA
Classement général458
31.0
AA
Science499
27.0
AA

Scores de benchmarks (LLM Stats)

(LLM Stats (zeroeval))

Chat

IFEvalGoogle Research (2023)88.6%Aut.

General

BFCL88.5%Aut.
MMLU87.3%Aut.
Multipl-E MBPP65.7%Aut.
Nexus58.7%Aut.

Language

MMLU (CoT)88.6%Aut.
Multipl-E HumanEval75.2%Aut.
MMLU-Pro73.3%Aut.

Math

GSM8k96.8%Aut.
Multilingual MGSM (CoT)91.6%Aut.
MATH73.8%Aut.

Reasoning

ARC-C96.9%Aut.
API-Bank92.0%Aut.
HumanEvalOpenAI (2021)89.0%Aut.
MBPP EvalPlus88.6%Aut.
DROP84.8%Aut.
GPQANYU + Cohere + Anthropic (2023)50.7%Aut.
Gorilla Benchmark API Bench35.3%Aut.

Indices d'évaluation AA

(Artificial Analysis)
Mmlu Pro(TIGER-Lab (Univ. of Waterloo, Toronto, CMU, 2024))
73.2
Math 500(OpenAI (2024), subset of Hendrycks et al. MATH (2021))
70.3
Gpqa(NYU + Cohere + Anthropic (2023))
51.5
Ifbench(Google Research (2023))
39.0
Livecodebench(UC Berkeley + MIT + Cornell (2024))
30.5
Lcr(Artificial Analysis)
25.3
Aime(MAA (Mathematical Association of America))
21.3
Tau2(Sierra + U Toronto + Vector Institute (2025))
19.0
Intelligence Index(Artificial Analysis)
7.3
Terminalbench Hard(Stanford × Laude Institute (2026))
6.8
Hle(Center for AI Safety + Scale AI (2025))
4.0
Math Index(Artificial Analysis)
3.0
Aime 25(MAA (Mathematical Association of America))
3.0

Scores par catégorie LLM Stats

(LLM Stats (zeroeval))
Chat
90
Instruction Following
90
Math
90
Structured Output
90
Language
80
Legal
80
Reasoning
80
Finance
80
General
80
Healthcare
80
Tool Calling
70
Code
60
Physics
50
Biology
50
Chemistry
50

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

Sources externes

LLM StatsArtificial Analysis