Llama 3.1 405B Instruct
Zero-eval
#1ARC-C
#1API-Bank
#1Multilingual MGSM (CoT)
+7 more
by Meta
About
Llama 3.1 405B Instruct is a language model developed by Meta. It achieves strong performance with an average score of 79.2% across 18 benchmarks. It excels particularly in ARC-C (96.9%), GSM8k (96.8%), API-Bank (92.0%). It supports a 256K token context window for handling large documents. The model is available through 8 API providers. Released in 2024, it represents Meta's latest advancement in AI technology.
Pricing Range
Input (per 1M)$0.89 -$9.50
Output (per 1M)$0.89 -$16.00
Providers8
Timeline
AnnouncedJul 23, 2024
ReleasedJul 23, 2024
Specifications
Training Tokens15.0T
License & Family
License
Llama 3.1 Community License
Benchmark Performance Overview
Performance metrics and category breakdown
Overall Performance
18 benchmarks
Average Score
79.2%
Best Score
96.9%
High Performers (80%+)
11Performance Metrics
Max Context Window
256.0KAvg Throughput
48.3 tok/sAvg Latency
0msTop Categories
reasoning
96.9%
math
87.4%
code
81.4%
general
73.2%
Benchmark Performance
Top benchmark scores with normalized values (0-100%)
Ranking Across Benchmarks
Position relative to other models on each benchmark
ARC-C
Rank #1 of 31
#1Llama 3.1 405B Instruct
96.9%
#2Claude 3 Opus
96.4%
#3Nova Pro
94.8%
#4Llama 3.1 70B Instruct
94.8%
GSM8k
Rank #4 of 46
#1GPT-4.5
97.0%
#2o1
97.1%
#3Kimi K2 Instruct
97.3%
#4Llama 3.1 405B Instruct
96.8%
#5Claude 3.5 Sonnet
96.4%
#6Claude 3.5 Sonnet
96.4%
#7Gemma 3 27B
95.9%
API-Bank
Rank #1 of 3
#1Llama 3.1 405B Instruct
92.0%
#2Llama 3.1 70B Instruct
90.0%
#3Llama 3.1 8B Instruct
82.6%
Multilingual MGSM (CoT)
Rank #1 of 3
#1Llama 3.1 405B Instruct
91.6%
#2Llama 3.1 70B Instruct
86.9%
#3Llama 3.1 8B Instruct
68.9%
HumanEval
Rank #13 of 62
#10Gemini Diffusion
89.6%
#11Granite 3.3 8B Base
89.7%
#12Granite 3.3 8B Instruct
89.7%
#13Llama 3.1 405B Instruct
89.0%
#14Nova Pro
89.0%
#15DeepSeek-V2.5
89.0%
#16Mistral Small 3.1 24B Instruct
88.4%
All Benchmark Results for Llama 3.1 405B Instruct
Complete list of benchmark scores with detailed information
ARC-C ARC-C benchmark | reasoning | text | 0.97 | 96.9% | Self-reported |
GSM8k GSM8k benchmark | math | text | 0.97 | 96.8% | Self-reported |
API-Bank API-Bank benchmark | general | text | 0.92 | 92.0% | Self-reported |
Multilingual MGSM (CoT) Multilingual MGSM (CoT) benchmark | math | text | 0.92 | 91.6% | Self-reported |
HumanEval HumanEval benchmark | code | text | 0.89 | 89.0% | Self-reported |
MMLU (CoT) MMLU (CoT) benchmark | general | text | 0.89 | 88.6% | Self-reported |
IFEval IFEval benchmark | code | text | 0.89 | 88.6% | Self-reported |
MBPP EvalPlus MBPP EvalPlus benchmark | code | text | 0.89 | 88.6% | Self-reported |
BFCL BFCL benchmark | general | text | 0.89 | 88.5% | Self-reported |
MMLU MMLU benchmark | general | text | 0.87 | 87.3% | Self-reported |
Showing 1 to 10 of 18 benchmarks