Key capabilities

About this model

The Llama 3 instruction tuned models are optimized for dialogue use cases and outperform many of the available open source chat models on common industry benchmarks. Instruction tuned models are intended for assistant-like chat, whereas pretrained models can be adapted for a variety of natural language generation tasks.

Key model capabilities

In this section, we report the results for Llama 3 models on standard automatic benchmarks. For all the evaluations, we use our internal evaluations library. For details on the methodology see here .

Base pretrained models

Category	Benchmark	Llama 3 8B	Llama2 7B	Llama2 13B	Llama 3 70B	Llama2 70B
General	MMLU (5-shot)	66.6	45.7	53.8	79.5	69.7
	AGIEval English (3-5 shot)	45.9	28.8	38.7	63.0	54.8
	CommonSenseQA (7-shot)	72.6	57.6	67.6	83.8	78.7
	Winogrande (5-shot)	76.1	73.3	75.4	83.1	81.8
	BIG-Bench Hard (3-shot, CoT)	61.1	38.1	47.0	81.3	65.7
	ARC-Challenge (25-shot)	78.6	53.7	67.6	93.0	85.3
Knowledge reasoning	TriviaQA-Wiki (5-shot)	78.5	72.1	79.6	89.7	87.5
Reading comprehension	SQuAD (1-shot)	76.4	72.2	72.1	85.6	82.6
	QuAC (1-shot, F1)	44.4	39.6	44.9	51.1	49.4
	BoolQ (0-shot)	75.7	65.5	66.9	79.0	73.1
	DROP (3-shot, F1)	58.4	37.9	49.8	79.7	70.2

Instruction tuned models

Benchmark	Llama 3 8B	Llama 2 7B	Llama 2 13B	Llama 3 70B	Llama 2 70B
MMLU (5-shot)	68.4	34.1	47.8	82.0	52.9
GPQA (0-shot)	34.2	21.7	22.3	39.5	21.0
HumanEval (0-shot)	62.2	7.9	14.0	81.7	25.6
GSM-8K (8-shot, CoT)	79.6	25.7	77.4	93.0	57.5
MATH (4-shot, CoT)	30.0	3.8	6.7	50.4	11.6

Use cases

See Responsible AI for additional considerations for responsible use.

Key use cases

Llama 3 is intended for commercial and research use in English. Instruction tuned models are intended for assistant-like chat, whereas pretrained models can be adapted for a variety of natural language generation tasks. Enables applications to be Built with Meta Llama 3.

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 Community License. Use in languages other than English**. **Note: Developers may fine-tune Llama 3 models for languages beyond English provided they comply with the Llama

Pricing

Pricing is based on a number of factors, including deployment type and tokens used. See pricing details here.

Technical specs

Llama 3 is an auto-regressive language model that uses an optimized transformer architecture. The tuned versions use supervised fine-tuning (SFT) and reinforcement learning with human feedback (RLHF) to align with human preferences for helpfulness and safety. Llama 3 comes in two sizes — 8B and 70B parameters — in pre-trained and instruction tuned variants.

	Training Data	Params	Context length	GQA	Token count	Knowledge cutoff
Llama 3	A new mix of publicly available online data.	8B	8k	Yes	15T+	March, 2023
Llama 3	A new mix of publicly available online data.	70B	8k	Yes	15T+	December, 2023

Both the 8 and 70B versions use Grouped-Query Attention (GQA) for improved inference scalability.

Training cut-off date

The pretraining data has a cutoff of March 2023 for the 8B and December 2023 for the 70B models respectively.

Training time

Pretraining utilized a cumulative 7.7M GPU hours of computation on hardware of type H100-80GB (TDP of 700W). Estimated total emissions were 2290 tCO2eq, 100% of which were offset by Meta's sustainability program.

	Time (GPU hours)	Power Consumption (W)	Carbon Emitted(tCO2eq)
Llama 3 8B	1.3M	700	390
Llama 3 70B	6.4M	700	1900
Total	7.7M		2290

We used custom training libraries, Meta's Research SuperCluster, and production clusters for pretraining. Fine-tuning, annotation, and evaluation were also performed on third-party cloud compute.

Input formats

Models input text only.

Output formats

Models generate text and code only.

Supported languages

Llama 3 is intended for commercial and research use in English.

Sample JSON response

Sample input

{
  "input_data": {
      "input_string": ["I believe the meaning of life is"],
      "parameters":{   
              "top_p": 0.8,
              "temperature": 0.8,
              "max_new_tokens": 100,
              "do_sample": true
      }
  }
}

Sample output

[
    {
        "0": "I believe the meaning of life is to be happy. I think we should always strive to be happy and live our lives to the fullest. I don't think we should always worry about what other people think. We should always do what makes us happy and what we think is right. I think it's important to always be yourself and never try to be someone you're not. I think it's important to always be positive and never give up. I think it's important to always believe in yourself and never let anyone tell you that"
    }
]

Model architecture

Llama 3 is an auto-regressive language model that uses an optimized transformer architecture. The tuned versions use supervised fine-tuning (SFT) and reinforcement learning with human feedback (RLHF) to align with human preferences for helpfulness and safety.

Long context

Both the 8 and 70B versions have 8k context length.

Optimizing model performance

The provider has not supplied this information.

Additional assets

For more technical information about generation parameters and recipes for how to use Llama 3 in applications, please go here . Instructions on how to provide feedback or comments on the model can be found in the model README . We provide a set of resources including Meta Llama Guard 2 and Code Shield safeguards. These tools have proven to drastically reduce residual risks of LLM Systems, while maintaining a high level of helpfulness. We encourage developers to tune and deploy these safeguards according to their needs and we provide a reference implementation to get you started. 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 .

Training disclosure

Training, testing and validation

Llama 3 was pretrained on over 15 trillion tokens of data from publicly available sources. The fine-tuning data includes publicly available instruction datasets, as well as over 10M human-annotated examples. Neither the pretraining nor the fine-tuning datasets include Meta user data.

Responsible AI considerations

Safety techniques

For our instruction tuned model, we conducted extensive red teaming exercises, performed adversarial evaluations and implemented safety mitigations techniques to lower residual risks. As with any Large Language Model, residual risks will likely remain and we recommend that developers assess these risks in the context of their use case. In parallel, we are working with the community to make AI safety benchmark standards transparent, rigorous and interpretable. The Llama 3 instruction tuned models are optimized for dialogue use cases and outperform many of the available open source chat models on common industry benchmarks. Further, in developing these models, we took great care to optimize helpfulness and safety. Llama 3 is an auto-regressive language model that uses an optimized transformer architecture. The tuned versions use supervised fine-tuning (SFT) and reinforcement learning with human feedback (RLHF) to align with human preferences for helpfulness and safety. In addition to residual risks, we put a great emphasis on model refusals to benign prompts. Over-refusing not only can impact the user experience but could even be harmful in certain contexts as well. We've heard the feedback from the developer community and improved our fine tuning to ensure that Llama 3 is significantly less likely to falsely refuse to answer prompts than Llama 2. We built internal benchmarks and developed mitigations to limit false refusals making Llama 3 our most helpful model to date. As part of the Llama 3 release, we updated our Responsible Use Guide to outline the steps and best practices for developers to implement model and system level safety for their application. We also provide a set of resources including Meta Llama Guard 2 and Code Shield safeguards. These tools have proven to drastically reduce residual risks of LLM Systems, while maintaining a high level of helpfulness. We encourage developers to tune and deploy these safeguards according to their needs and we provide a reference implementation to get you started.

Safety evaluations

We have conducted a two fold assessment of the safety of the model in this area:

Testing against a benchmark combining CBRNE and adversarial intent, as well as fine tuning the model to help ensure it refuses to provide detailed information to promote potential CBRNE harm.
Involving external CBRNE experts to conduct an uplift test assessing the ability of the model to accurately provide expert knowledge and reduce barriers to potential CBRNE misuse, by reference to what can be achieved using web search (without the model).

We have evaluated Llama 3 with CyberSecEval, Meta's cybersecurity safety eval suite, measuring Llama 3's propensity to suggest insecure code when used as a coding assistant, and Llama 3's propensity to comply with requests to help carry out cyber attacks, where attacks are defined by the industry standard MITRE ATT&CK cyber attack ontology. On our insecure coding and cyber attacker helpfulness tests, Llama 3 behaved in the same range or safer than models of equivalent coding capability. 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. We also partnered with content specialists to perform red teaming exercises assessing potentially violating content while taking account of market specific nuances or experiences.

Known limitations

The core values of Llama 3 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 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 is a new technology, and like any new technology, there are risks associated with its use. Testing conducted to date has been in English, and has not covered, nor could it cover, all scenarios. For these reasons, as with all LLMs, Llama 3'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 models, developers should perform safety testing and tuning tailored to their specific applications of the model. As outlined in the Responsible Use Guide, we recommend incorporating Purple Llama solutions into your workflows and specifically Llama Guard which provides a base model to filter input and output prompts to layer system-level safety on top of model-level safety. Foundation models are widely capable technologies that are built to be used for a diverse range of applications. They are not designed to meet every developer preference on safety levels for all use cases, out-of-the-box, as those by their nature will differ across different applications. Rather, responsible LLM-application deployment is achieved by implementing a series of safety best practices throughout the development of such applications, from the model pre-training, fine-tuning and the deployment of systems composed of safeguards to tailor the safety needs specifically to the use case and audience. As outlined in the Responsible Use Guide, some trade-off between model helpfulness and model alignment is likely unavoidable. Developers should exercise discretion about how to weigh the benefits of alignment and helpfulness for their specific use case and audience. Developers should be mindful of residual risks when using Llama models and leverage additional safety tools as needed to reach the right safety bar for their use case.

Acceptable use

Acceptable use policy

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 Community License. Use in languages other than English. Note: Developers may fine-tune Llama 3 models for languages beyond English provided they comply with the Llama 3 Community License and the Acceptable Use Policy. If you access or use Llama 3, you agree to the Acceptable Use Policy. The most recent copy of this policy can be found at https://llama.meta.com/llama3/use-policy/ .

Quality and performance evaluations

Source: Meta In this section, we report the results for Llama 3 models on standard automatic benchmarks. For all the evaluations, we use our internal evaluations library. For details on the methodology see here .

Base pretrained models

Category	Benchmark	Llama 3 8B	Llama2 7B	Llama2 13B	Llama 3 70B	Llama2 70B
General	MMLU (5-shot)	66.6	45.7	53.8	79.5	69.7
	AGIEval English (3-5 shot)	45.9	28.8	38.7	63.0	54.8
	CommonSenseQA (7-shot)	72.6	57.6	67.6	83.8	78.7
	Winogrande (5-shot)	76.1	73.3	75.4	83.1	81.8
	BIG-Bench Hard (3-shot, CoT)	61.1	38.1	47.0	81.3	65.7
	ARC-Challenge (25-shot)	78.6	53.7	67.6	93.0	85.3
Knowledge reasoning	TriviaQA-Wiki (5-shot)	78.5	72.1	79.6	89.7	87.5
Reading comprehension	SQuAD (1-shot)	76.4	72.2	72.1	85.6	82.6
	QuAC (1-shot, F1)	44.4	39.6	44.9	51.1	49.4
	BoolQ (0-shot)	75.7	65.5	66.9	79.0	73.1
	DROP (3-shot, F1)	58.4	37.9	49.8	79.7	70.2

Instruction tuned models

Benchmark	Llama 3 8B	Llama 2 7B	Llama 2 13B	Llama 3 70B	Llama 2 70B
MMLU (5-shot)	68.4	34.1	47.8	82.0	52.9
GPQA (0-shot)	34.2	21.7	22.3	39.5	21.0
HumanEval (0-shot)	62.2	7.9	14.0	81.7	25.6
GSM-8K (8-shot, CoT)	79.6	25.7	77.4	93.0	57.5
MATH (4-shot, CoT)	30.0	3.8	6.7	50.4	11.6

We have evaluated Llama 3 with CyberSecEval, Meta's cybersecurity safety eval suite, measuring Llama 3's propensity to suggest insecure code when used as a coding assistant, and Llama 3's propensity to comply with requests to help carry out cyber attacks, where attacks are defined by the industry standard MITRE ATT&CK cyber attack ontology. On our insecure coding and cyber attacker helpfulness tests, Llama 3 behaved in the same range or safer than models of equivalent coding capability . We have conducted a two fold assessment of the safety of the model in this area:

Testing against a benchmark combining CBRNE and adversarial intent, as well as fine tuning the model to help ensure it refuses to provide detailed information to promote potential CBRNE harm.
Involving external CBRNE experts to conduct an uplift test assessing the ability of the model to accurately provide expert knowledge and reduce barriers to potential CBRNE misuse, by reference to what can be achieved using web search (without the model).

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. We also partnered with content specialists to perform red teaming exercises assessing potentially violating content while taking account of market specific nuances or experiences. We built internal benchmarks and developed mitigations to limit false refusals making Llama 3 our most helpful model to date.

Benchmarking methodology

Source: Meta For our instruction tuned model, we conducted extensive red teaming exercises, performed adversarial evaluations and implemented safety mitigations techniques to lower residual risks. As with any Large Language Model, residual risks will likely remain and we recommend that developers assess these risks in the context of their use case. In parallel, we are working with the community to make AI safety benchmark standards transparent, rigorous and interpretable.

Public data summary

Source: Meta The provider has not supplied this information.