Compare Llama 3.1 405B Instruct vs. Llama 3.1 70B Instruct

Pricing About Us Blog Docs

Compare

to

Understand and compare Llama 3.1 405B Instruct vs. Llama 3.1 70B Instruct

Turn your documents into engaging podcast discussions.

Overview

Llama 3.1 405B Instruct was released the same time Llama 3.1 70B Instruct.

	Llama 3.1 405B Instruct	Llama 3.1 70B Instruct
Provider The entity that provides this model.	Meta	Meta
Input Context Window The number of tokens supported by the input context window.	128K tokens	128K tokens
Maximum Output Tokens The number of tokens that can be generated by the model in a single request.	2,048 tokens	2,048 tokens
Release Date When the model was first released.	July 23rd, 2024 4 months ago 2024-07-23	July 23rd, 2024 4 months ago 2024-07-23
Knowledge Cutoff Limit on the knowledge base used by the model.	December 2023 December 2023	December 2023 December 2023
Open Source	Yes	Yes
API Providers The providers that offer this model. (This is not an exhaustive list.)	Azure AI, AWS Bedrock, Google Cloud Vertex AI Model Garden, NVIDIA NIM, IBM watsonx	Azure AI, AWS Bedrock, Google Cloud Vertex AI Model Garden, NVIDIA NIM, IBM watsonx

Pricing

	Llama 3.1 405B Instruct	Llama 3.1 70B Instruct
Input Cost of input data provided to the model.	Pricing not available.	Pricing not available.
Output Cost of output tokens generated by the model.	Pricing not available.	Pricing not available.

Benchmarks

Compare relevant benchmarks between Llama 3.1 405B Instruct and Llama 3.1 70B Instruct.

	Llama 3.1 405B Instruct	Llama 3.1 70B Instruct
MMLU Evaluating LLM knowledge acquisition in zero-shot and few-shot settings.	85.2 (5-shot) Source	83.6 (5-shot) Source
MMMU A wide ranging multi-discipline and multimodal benchmark.	Benchmark not available.	Benchmark not available.
HellaSwag A challenging sentence completion benchmark.	Benchmark not available.	Benchmark not available.
GSM8K Grade-school math problems benchmark.	96.8 (8-shot) Source	95.1 (8-shot) Source
HumanEval A benchmark to measure functional correctness for synthesizing programs from docstrings.	Benchmark not available.	Benchmark not available.
MATH Benchmark performance on Math problems ranging across 5 levels of difficulty and 7 sub-disciplines.	73.8 (0-shot) Source	68.0 (0-shot) Source

Compare more models

Llama 3.1 405B Instruct vs.

Llama 3.1 405B Instruct vs.

Llama 3.1 405B Instruct vs.

GPT-3.5 Turbo 0125

Llama 3.1 405B Instruct vs.

GPT-3.5 Turbo 16K

Llama 3.1 70B Instruct vs.

Llama 3.1 70B Instruct vs.

Llama 3.1 70B Instruct vs.

Llama 3.1 70B Instruct vs.

Mistral Large 2

Measure & Improve LLM Product Performance.