Intended Use

Primary Use Cases

Llama 3.1 is intended for commercial and research use in multiple languages. Instruction tuned text only models are intended for assistant-like chat, whereas pretrained models can be adapted for a variety of natural language generation tasks. The Llama 3.1 model collection also supports the ability to leverage the outputs of its models to improve other models including synthetic data generation and distillation. The Llama 3.1 Community License allows for these use cases.

Out-of-Scope Use Cases

Use in any manner that violates applicable laws or regulations (including trade compliance laws). Use in any other way that is prohibited by the Acceptable Use Policy and Llama 3.1 Community License. Use in languages beyond those explicitly referenced as supported in this model card**. Note: Llama 3.1 has been trained on a broader collection of languages than the 10 supported languages. Developers may fine-tune Llama 3.1 models for languages beyond the 8 supported languages provided they comply with the Llama 3.1 Community License and the Acceptable Use Policy and in such cases are responsible for ensuring that any uses of Llama 3.1 in additional languages is done in a safe and responsible manner.

Responsible AI Considerations

As part of our Responsible release approach, we followed a three-pronged strategy to managing trust & safety risks: Enable developers to deploy helpful, safe and flexible experiences for their target audience and for the use cases supported by Llama. Protect developers against adversarial users aiming to exploit Llama capabilities to potentially cause harm. Provide protections for the community to help prevent the misuse of our models. Responsible Deployment Llama is a foundational technology designed to be used in a variety of use cases, examples on how Meta's Llama models have been responsibly deployed can be found in our Community Stories webpage. Our approach is to build the most helpful models enabling the world to benefit from the technology power, by aligning our model safety for the generic use cases addressing a standard set of harms. Developers are then in the driver seat to tailor safety for their use case, defining their own policy and deploying the models with the necessary safeguards in their Llama systems. Llama 3.1 was developed following the best practices outlined in our Responsible Use Guide, you can refer to the Responsible Use Guide to learn more. Llama 3.1 Instruct Our main objectives for conducting safety fine-tuning are to provide the research community with a valuable resource for studying the robustness of safety fine-tuning, as well as to offer developers a readily available, safe, and powerful model for various applications to reduce the developer workload to deploy safe AI systems. For more details on the safety mitigations implemented please read the Llama 3 paper. Fine-Tuning Data We employ a multi-faceted approach to data collection, combining human-generated data from our vendors with synthetic data to mitigate potential safety risks. We've developed many large language model (LLM)-based classifiers that enable us to thoughtfully select high-quality prompts and responses, enhancing data quality control. Refusals And Tone Building on the work we started with Llama 3, we put a great emphasis on model refusals to benign prompts as well as refusal tone. We included both borderline and adversarial prompts in our safety data strategy, and modified our safety data responses to follow tone guidelines. Evaluations We evaluated Llama models for common use cases as well as specific capabilities. Common use cases evaluations measure safety risks of systems for most commonly built applications including chat bot, coding assistant, tool calls. We built dedicated, adversarial evaluation datasets and evaluated systems composed of Llama models and Llama Guard 3 to filter input prompt and output response. It is important to evaluate applications in context, and we recommend building a dedicated evaluation dataset for your use case. Prompt Guard and Code Shield are also available if relevant to the application. Capability evaluations measure vulnerabilities of Llama models inherent to specific capabilities, for which were crafted dedicated benchmarks including long context, multilingual, tools calls, coding or memorization. Red Teaming For both scenarios, we conducted recurring red teaming exercises with the goal of discovering risks via adversarial prompting and we used the learnings to improve our benchmarks and safety tuning datasets. We partnered early with subject-matter experts in critical risk areas to understand the nature of these real-world harms and how such models may lead to unintended harm for society. Based on these conversations, we derived a set of adversarial goals for the red team to attempt to achieve, such as extracting harmful information or reprogramming the model to act in a potentially harmful capacity. The red team consisted of experts in cybersecurity, adversarial machine learning, responsible AI, and integrity in addition to multilingual content specialists with background in integrity issues in specific geographic markets. Critical And Other Risks We specifically focused our efforts on mitigating the following critical risk areas:

• CBRNE (Chemical, Biological, Radiological, Nuclear, and Explosive materials) helpfulness

To assess risks related to proliferation of chemical and biological weapons, we performed uplift testing designed to assess whether use of Llama 3.1 models could meaningfully increase the capabilities of malicious actors to plan or carry out attacks using these types of weapons.

• Child Safety

Child Safety risk assessments were conducted using a team of experts, to assess the model's capability to produce outputs that could result in Child Safety risks and inform on any necessary and appropriate risk mitigations via fine tuning. We leveraged those expert red teaming sessions to expand the coverage of our evaluation benchmarks through Llama 3 model development. For Llama 3, we conducted new in-depth sessions using objective based methodologies to assess the model risks along multiple attack vectors including the additional languages Llama 3 is trained on. We also partnered with content specialists to perform red teaming exercises assessing potentially violating content while taking account of market specific nuances or experiences.

• Cyber Attack Enablement

Our cyber attack uplift study investigated whether LLMs can enhance human capabilities in hacking tasks, both in terms of skill level and speed. Our attack automation study focused on evaluating the capabilities of LLMs when used as autonomous agents in cyber offensive operations, specifically in the context of ransomware attacks. This evaluation was distinct from previous studies that considered LLMs as interactive assistants. The primary objective was to assess whether these models could effectively function as independent agents in executing complex cyber-attacks without human intervention. Our study of Llama-3.1-405B's social engineering uplift for cyber attackers was conducted to assess the effectiveness of AI models in aiding cyber threat actors in spear phishing campaigns. Please read our Llama 3.1 Cyber security whitepaper to learn more.

• Community

Generative AI safety requires expertise and tooling, and we believe in the strength of the open community to accelerate its progress. We are active members of open consortiums, including the AI Alliance, Partnership on AI and MLCommons, actively contributing to safety standardization and transparency. We encourage the community to adopt taxonomies like the MLCommons Proof of Concept evaluation to facilitate collaboration and transparency on safety and content evaluations. Our Purple Llama tools are open sourced for the community to use and widely distributed across ecosystem partners including cloud service providers. We encourage community contributions to our Github repository. We also set up the Llama Impact Grants program to identify and support the most compelling applications of Meta's Llama model for societal benefit across three categories: education, climate and open innovation. The 20 finalists from the hundreds of applications can be found here. Finally, we put in place a set of resources including an output reporting mechanism and bug bounty program to continuously improve the Llama technology with the help of the community.

• Ethical Considerations And Limitations

The core values of Llama 3.1 are openness, inclusivity and helpfulness. It is meant to serve everyone, and to work for a wide range of use cases. It is thus designed to be accessible to people across many different backgrounds, experiences and perspectives. Llama 3.1 addresses users and their needs as they are, without insertion unnecessary judgment or normativity, while reflecting the understanding that even content that may appear problematic in some cases can serve valuable purposes in others. It respects the dignity and autonomy of all users, especially in terms of the values of free thought and expression that power innovation and progress. But Llama 3.1 is a new technology, and like any new technology, there are risks associated with its use. Testing conducted to date has not covered, nor could it cover, all scenarios. For these reasons, as with all LLMs, Llama 3.1's potential outputs cannot be predicted in advance, and the model may in some instances produce inaccurate, biased or other objectionable responses to user prompts. Therefore, before deploying any applications of Llama 3.1 models, developers should perform safety testing and tuning tailored to their specific applications of the model. Please refer to available resources including our Responsible Use Guide, Trust and Safety solutions, and other resources to learn more about responsible development.

Training Data

Overview: Llama 3.1 was pretrained on ~15 trillion tokens of data from publicly available sources. The fine-tuning data includes publicly available instruction datasets, as well as over 25M synthetically generated examples. Data Freshness: The pretraining data has a cutoff of December 2023.

The Llama 3.1 8B Instruct NIM microservice has the same evaluation scores as the original Meta Llama 3.1 8B model. The NIM microservice optimizes the model to run with best throughput and lowest latency on NVIDIA GPU compute instances on Azure AI Foundry.

Model	MMLU (5 shots) macro_avg/acc	MMLU (CoT) (0 shots) macro_avg/acc	MMLU PRO (CoT) (5 shots) micro_avg/acc_char	ARC-C (0 shots) acc	GPQA (0 shots) em	MuSR (0 shots) correct	IFEval	HumanEval (0 shots) pass@1	MBPP ++ base version (0 shots) pass@1	GSM-8K (CoT) (8 shots) em_maj1@1	MATH (CoT) (0 shots) final_em	API-Bank (0 shots) acc	Berkeley Function Calling (0 shots) acc	Gorilla Benchmark API Bench (0 shots) acc	Nexus (0-shot) macro_avg/acc	Multilingual MGSM (8 shots) em
Llama 3 8B Instruct	68.5	65.3	45.5	82.4	34.6	56.3	76.8	60.4	70.6	80.6	29.1	83.6	76.1	8.8	37.6	-
Llama 3.1 8B Instruct	69.4	72.7	48.3	83.4	30.4	45.7	80.4
72.6	72.8	84.5	51.9	82.6	76.1	8.2	38.5	68.2
Llama 3 70B Instruct	82	80.9	63.4	94.4	39.5	55.1	82.9	81.7	82.5	93	51	85.1	83	14.7	47.8	-
Llama 3.1 70B Instruct	83.6	85.9	65.1	94.8	41.7	58.1	87.5	80.5	86	95.1	68	90	85.1	29.7	56.7	85.6
Llama 3.1 405B Instruct	87.3	88.6	73.3	96.9	50.7	56.7	88.6	89	88.6	96.8	73.8	92	88.5	35.3	58.7	90.3
For more information on evaluation benchmarks, please refer [here] (https://build.nvidia.com/meta/llama-3_1-8b-instruct/modelcard )

Llama 3.1 8B NIM is optimized to run best on the following compute:

GPU	Total GPU memory	Azure VM compute	#GPUs on VM	Link
A100	80	Standard_NC24ads_A100_v4	1	link
A100	160	Standard_NC48ads_A100_v4	2	link
A100	320	Standard_NC96ads_A100_v4	4	link
A100	640	STANDARD_ND96AMSR_A100_V4	8	link
H100	94	STANDARD_NC40ADS_H100_V5	1	link
H100	188	STANDARD_NC80ADIS_H100_V5	2	link