The Llama 3 instruction tuned models are optimized for dialogue use cases and outperform many of the available open source chat llm on common industry benchmarks
Meta developed and released the Meta Llama 3 family of large language models (LLMs), a collection of pretrained and instruction tuned generative text models in 8 and 70B sizes. 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.
Model developers Meta
Input Models input text only.
Output Models generate text and code only.
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.
What's new with Llama-3?
Llama 3 brings significant enhancements over Llama 2, including a new tokenizer that increases the vocabulary size to 128,256 tokens (up from 32K tokens). This expanded vocabulary enhances text encoding efficiency, promoting stronger multilingual capabilities.
Moreover, Llama 3 models underwent extensive training on a diverse dataset comprising over 15 trillion tokens, approximately eight times more data than its predecessor. Specifically, Llama 3 Instruct, tailored for dialogue applications, was fine-tuned on a dataset of over 10 million human-annotated samples using a combination of techniques such as supervised fine-tuning, rejection sampling, proximal policy optimization, and direct policy optimization.
Llama-3-8B-Instruct
Training Data
Params
Context length
GQA
Token count
Knowledge cutoff
A new mix of publicly available online data.
8B
8k
Yes
15T+
March, 2023
Token counts refer to pretraining data only. Both the 8 and 70B versions use Grouped-Query Attention (GQA) for improved inference scalability.
Prompt template
The base models have no prompt format. Like other base models, they can be used to continue an input sequence with a plausible continuation or for zero-shot/few-shot inference. They are also a great foundation for fine-tuning your own use cases. The Instruct versions use the following conversation structure:
Llama-3-Instruct is a highly capable AI assistant that excels in complex reasoning tasks across various domains. It can be used for a wide range of applications, including but not limited to:
1. Conversational Agents:
Input: "Hi there! I need some advice on planning a trip to Japan. Can you help me with the best places to visit?"
Output: "
Japan! What an amazing country! I'd be delighted to help you plan your trip. Japan has so much to offer, from vibrant cities to serene landscapes, rich culture, and delicious food. Here are some of the must-visit places to consider:
**Cities:**
1. Tokyo: The capital city is a must-visit, with its neon-lit skyscrapers, bustling streets, and world-class restaurants. Visit the famous Shibuya Crossing, Tokyo Tower, and the Meiji Shrine.
2. Kyoto: This city is steeped in tradition and history, with numerous temples, shrines, and gardens. Don't miss the Fushimi Inari Shrine, Kinkaku-ji Temple, and Arashiyama Bamboo Grove.
3. Osaka: Known for its food, entertainment, and nightlife, Osaka is a great place to try local cuisine like takoyaki and okonomiyaki.
4. Hiroshima: Visit the Hiroshima Peace Memorial Park and Museum to learn about the city's history and the atomic bombing in 1945.
**Nature and Landscapes:**
1. Mount Fuji: Take a day trip to the Fuji Five Lakes region or climb the mountain for a breathtaking view.
2. Nikko: Located about two hours from Tokyo, Nikko is famous for its ornate temples and shrines, including the Toshogu Shrine, a UNESCO World Heritage Site.
3. Japanese Alps: Visit the Nagano or Niigata prefectures for stunning mountain scenery, hot springs, and ski resorts.
4. Okinawa: Head to the southernmost islands for beautiful beaches, crystal-clear waters, and a unique Ryukyu culture.
**Other attractions:**
1. Miyajima Island: Famous for its scenic beauty and historic landmarks, including the famous Itsukushima Shrine, which appears to be floating during high tide.
2. Nara: Visit the ancient city of Nara, with its friendly deer, temples, and shrines.
3. Hakone: Take a scenic ride on the Hakone Ropeway for stunning views of Mt. Fuji and Lake Ashi.
4. Onsen (Hot Springs): Japan is famous for its natural hot springs, which are perfect for relaxing after a long day of sightseeing.
**Tips and Essentials:**
1. Language: While many Japanese people speak some English, it's still a good idea to learn basic Japanese phrases and carry a translation guide.
2. Transportation: Japan has an efficient public transportation system, including trains, buses"
Input: "Hey, I'm feeling a bit down today. Can you suggest some ways to improve my mood?"
Output:
I'm so sorry to hear that you're feeling down! Here are some suggestions that might help improve your mood:
1. Take a short walk outside and get some fresh air.
2. Listen to your favorite music or podcast.
3. Practice some deep breathing exercises or meditation.
4. Reach out to a friend or family member and have a chat.
5. Do something creative, like drawing or writing.
Remember, it's okay to feel down sometimes, and it's important to take care of yourself. If you're feeling overwhelmed or struggling with persistent feelings of sadness, it's always a good idea to reach out to a mental health professional for support."
Note: The output is a friendly and empathetic response that acknowledges the user's feelings and offers suggestions for improving their mood. The tone is warm and non-judgmental, and the language is clear and easy to understand. The response also includes a gentle reminder that it's okay to feel down sometimes and that seeking professional help is always an option.
2. Mental Health
The Llama-3-Instruct model can be used to build chatbots that can provide mental health support to individuals.
Input: "I'm struggling with depression and finding it hard to stay motivated. Can you suggest some ways to improve my mood?"
Output: "I'm struggling with depression and finding it hard to stay motivated. Can you suggest some ways to improve my mood? Here are a few suggestions:
1. **Exercise**: Regular physical activity can help reduce symptoms of depression and improve mood. Try going for a walk, jog, or bike ride, or join a fitness class.
2. **Connect with others**: Social support is crucial for mental health. Reach out to friends, family, or a therapist for a chat or to schedule a meetup.
3. **Practice self-care**: Take care of your physical and emotional needs by getting enough sleep, eating a balanced diet, and engaging in activities that bring you joy.
4. **Challenge negative thoughts**: When you notice negative thoughts creeping in, try to reframe them in a more positive or realistic light. For example, instead of thinking "I'll never be able to do this," try reframing it as "I'll learn and grow from this experience."
5. **Seek professional help**: If you're struggling with depression, it's important to seek help from a mental health professional. They can provide you with personalized guidance and support to help you manage your symptoms and improve your mood.
Remember, improving your mood takes time and effort, but it's possible with the right strategies and support. Don't be afraid to reach out for help when you need it."
Note: The output is a suggested response and may not be a perfect match for the input. The goal is to provide a helpful and supportive response that acknowledges the user's struggles and offers practical suggestions for improving their mood."
Input: "I've been feeling really overwhelmed and anxious lately. Can you help me cope with stress?"
Output:
"I understand that you're feeling overwhelmed and anxious. It sounds like you're experiencing a lot of stress. Can you tell me more about what's been going on and how you've been coping so far?"
This response acknowledges the person's feelings and shows empathy, while also encouraging them to share more information and take a more active role in the conversation. It also sets the stage for a collaborative problem-solving approach, where the therapist and client work together to develop strategies for managing stress and anxiety.
Here are some key elements of this response:
1. Acknowledgment: The response acknowledges the person's feelings and experiences, showing that the therapist is actively listening and trying to understand their perspective.
2. Empathy: The response expresses empathy and understanding, which can help the person feel heard and validated.
3. Open-ended question: The response includes an open-ended question that encourages the person to share more information and take a more active role in the conversation.
4. Focus on coping: The response focuses on the person's coping strategies and how they've been managing stress and anxiety, which can help the therapist understand their strengths and challenges.
5. Collaborative approach: The response sets the stage for a collaborative problem-solving approach, where the therapist and client work together to develop strategies for managing stress and anxiety.
Overall, this response is an example of a supportive and non-judgmental approach to therapy, which can help build trust and rapport with the client. It also sets the stage for a productive and collaborative conversation about managing stress and anxiety."
3. Education
Education: The Llama-3-Instruct model can be used to build chatbots that can assist students in learning and understanding various concepts.
Input: "Can you explain the concept of photosynthesis in a simple and easy-to-understand way?"
Output:
"Photosynthesis is the process by which plants, algae, and some bacteria convert light energy from the sun into chemical energy in the form of glucose. This process occurs in specialized organelles called chloroplasts, which contain the pigment chlorophyll. Chlorophyll absorbs light energy and uses it to power a series of chemical reactions that convert carbon dioxide and water into glucose and oxygen. The glucose produced during photosynthesis is used by the plant to fuel its metabolic processes, while the oxygen is released into the atmosphere as a byproduct. In simple terms, photosynthesis is like a magic trick where plants take in sunlight, water, and carbon dioxide and produce glucose and oxygen as a result.""
Note: The output is a rewritten version of the input text, rephrased to be more concise and easy to understand. The goal is to make the concept of photosynthesis accessible to a general audience."
Input: "I don't quite understand the quadratic equation. Can you walk me through it step by step?"
Output:
"I'd be happy to help you understand the quadratic equation. Here's a step-by-step explanation:
1. Start with the general form of the quadratic equation: ax^2 + bx + c = 0
2. Identify the values of a, b, and c
3. Use the quadratic formula: x = (-b ± √(b^2 - 4ac)) / 2a
4. Simplify the equation by combining like terms
5. Solve for x
Let me know if you have any questions or need further clarification!""
This model uses a combination of natural language processing (NLP) and machine learning algorithms to generate human-like responses to user input. The model is trained on a large dataset of text and can be fine-tuned for specific domains or topics. In this case, the model is trained on a dataset of math-related text and can generate responses to user input related to math problems. The model can also be used to generate responses to other types of user input, such as questions or statements about a particular topic."
4. Personal Assistants
Llama-3-Instruct model can be used to build personal assistants that can perform tasks, such as setting reminders, scheduling appointments, and sending messages
Input: "Remind me to attend the team meeting tomorrow at 2 PM."
Output: "I'd be happy to remind you. I'll send you a reminder notification tomorrow at 1:45 PM to attend the team meeting at 2 PM. Would you like me to add any specific details or tasks to your reminder, such as the meeting location or agenda items?"
Input: "Send a message to John saying I'll meet him for lunch at the usual spot."
Output:
'Here is the message: "Hey John, I'll meet you for lunch at the usual spot. See you then!"'
Training Data
Overview 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.
Data Freshness The pretraining data has a cutoff of March 2023 for the 8B and December 2023 for the 70B models respectively.
Hardware and Software
Training Factors 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.
Carbon Footprint 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
CO2 emissions during pre-training. Time: total GPU time required for training each model. Power Consumption: peak power capacity per GPU device for the GPUs used adjusted for power usage efficiency. 100% of the emissions are directly offset by Meta's sustainability program, and because we are openly releasing these models, the pretraining costs do not need to be incurred by others.
Intended Use
Intended 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.
Out-of-scope 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.
Benchmarks
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
Responsibility & Safety
We believe that an open approach to AI leads to better, safer products, faster innovation, and a bigger overall market. We are committed to Responsible AI development and took a series of steps to limit misuse and harm and support the open source community.
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 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.
Llama 3-Instruct
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.
Safety
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.
Refusals
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.
Responsible release
In addition to responsible use considerations outlined above, we followed a rigorous process that requires us to take extra measures against misuse and critical risks before we make our release decision.
Misuse
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/.
Critical risks
CBRNE (Chemical, Biological, Radiological, Nuclear, and high yield Explosives)
We have conducted a two fold assessment of the safety of the model in this area:
Iterative testing during model training to assess the safety of responses related to CBRNE threats and other adversarial risks.
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).
Cyber Security
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
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.
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 in 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.
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 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.
Aaditya Singh; Aaron Grattafiori; Abhimanyu Dubey; Abhinav Jauhri; Abhinav Pandey; Abhishek Kadian; Adam Kelsey; Adi Gangidi; Ahmad Al-Dahle; Amit Sangani; Ahuva Goldstand; Aiesha Letman; Ajay Menon; Akhil Mathur; Alan Schelten; Alex Vaughan; Amy Yang; Andrei Lupu; Andres Alvarado; Andrew Gallagher; Andrew Gu; Andrew Ho; Andrew Poulton; Andrew Ryan; Angela Fan; Ankit Ramchandani; Anthony Hartshorn; Archi Mitra; Archie Sravankumar; Artem Korenev; Arun Rao; Ashley Gabriel; Ashwin Bharambe; Assaf Eisenman; Aston Zhang; Ash JJhaveri; Aurelien Rodriguez; Austen Gregerson; Ava Spataru; Baptiste Roziere; Ben Maurer; Benjamin Leonhardi; Bernie Huang; Bhargavi Paranjape; Bing Liu; Binh Tang; Bobbie Chern; Brani Stojkovic; Brian Fuller; Catalina Mejia Arenas; Chao Zhou; Charlotte Caucheteux; Chaya Nayak; Ching-Hsiang Chu; Chloe Bi; Chris Cai; Chris Cox; Chris Marra; Chris McConnell; Christian Keller; Christoph Feichtenhofer; Christophe Touret; Chunyang Wu; Corinne Wong; Cristian Canton Ferrer; Damien Allonsius; Daniel Kreymer; Daniel Haziza; Daniel Li; Danielle Pintz; Danny Livshits; Danny Wyatt; David Adkins; David Esiobu; David Xu; Davide Testuggine; Delia David; Devi Parikh; Dhruv Choudhary; Dhruv Mahajan; Diana Liskovich; Diego Garcia-Olano; Diego Perino; Dieuwke Hupkes; Dingkang Wang; Dustin Holland; Egor Lakomkin; Elina Lobanova; Xiaoqing Ellen Tan; Emily Dinan; Eric Smith; Erik Brinkman; Esteban Arcaute; Filip Radenovic; Firat Ozgenel; Francesco Caggioni; Frank Seide; Frank Zhang; Gabriel Synnaeve; Gabriella Schwarz; Gabrielle Lee; Gada Badeer; Georgia Anderson; Graeme Nail; Gregoire Mialon; Guan Pang; Guillem Cucurell; Hailey Nguyen; Hamid Shojanazeri; Hannah Korevaar; Hannah Wang; Haroun Habeeb; Harrison Rudolph; Henry Aspegren; Hu Xu; Hugo Touvron; Iga Kozlowska; Igor Molybog; Igor Tufanov; Iliyan Zarov; Imanol Arrieta Ibarra; Irina-Elena Veliche; Isabel Kloumann; Ishan Misra; Ivan Evtimov; Jacob Xu; Jade Copet; Jake Weissman; Jan Geffert; Jana Vranes; Japhet Asher; Jason Park; Jay Mahadeokar; Jean-Baptiste Gaya; Jeet Shah; Jelmer van der Linde; Jennifer Chan; Jenny Hong; Jenya Lee; Jeremy Fu; Jeremy Teboul; Jianfeng Chi; Jianyu Huang; Jie Wang; Jiecao Yu; Joanna Bitton; Joe Spisak; Joelle Pineau; Jon Carvill; Jongsoo Park; Joseph Rocca; Joshua Johnstun; Junteng Jia; Kalyan Vasuden Alwala; Kam Hou U; Kate Plawiak; Kartikeya Upasani; Kaushik Veeraraghavan; Ke Li; Kenneth Heafield; Kevin Stone; Khalid El-Arini; Krithika Iyer; Kshitiz Malik; Kuenley Chiu; Kunal Bhalla; Kyle Huang; Lakshya Garg; Lauren Rantala-Yeary; Laurens van der Maaten; Lawrence Chen; Leandro Silva; Lee Bell; Lei Zhang; Liang Tan; Louis Martin; Lovish Madaan; Luca Wehrstedt; Lukas Blecher; Luke de Oliveira; Madeline Muzzi; Madian Khabsa; Manav Avlani; Mannat Singh; Manohar Paluri; Mark Zuckerberg; Marcin Kardas; Martynas Mankus; Mathew Oldham; Mathieu Rita; Matthew Lennie; Maya Pavlova; Meghan Keneally; Melanie Kambadur; Mihir Patel; Mikayel Samvelyan; Mike Clark; Mike Lewis; Min Si; Mitesh Kumar Singh; Mo Metanat; Mona Hassan; Naman Goyal; Narjes Torabi; Nicolas Usunier; Nikolay Bashlykov; Nikolay Bogoychev; Niladri Chatterji; Ning Dong; Oliver Aobo Yang; Olivier Duchenne; Onur Celebi; Parth Parekh; Patrick Alrassy; Paul Saab; Pavan Balaji; Pedro Rittner; Pengchuan Zhang; Pengwei Li; Petar Vasic; Peter Weng; Polina Zvyagina; Prajjwal Bhargava; Pratik Dubal; Praveen Krishnan; Punit Singh Koura; Qing He; Rachel Rodriguez; Ragavan Srinivasan; Rahul Mitra; Ramon Calderer; Raymond Li; Robert Stojnic; Roberta Raileanu; Robin Battey; Rocky Wang; Rohit Girdhar; Rohit Patel; Romain Sauvestre; Ronnie Polidoro; Roshan Sumbaly; Ross Taylor; Ruan Silva; Rui Hou; Rui Wang; Russ Howes; Ruty Rinott; Saghar Hosseini; Sai Jayesh Bondu; Samyak Datta; Sanjay Singh; Sara Chugh; Sargun Dhillon; Satadru Pan; Sean Bell; Sergey Edunov; Shaoliang Nie; Sharan Narang; Sharath Raparthy; Shaun Lindsay; Sheng Feng; Sheng Shen; Shenghao Lin; Shiva Shankar; Shruti Bhosale; Shun Zhang; Simon Vandenhende; Sinong Wang; Seohyun Sonia Kim; Soumya Batra; Sten Sootla; Steve Kehoe; Suchin Gururangan; Sumit Gupta; Sunny Virk; Sydney Borodinsky; Tamar Glaser; Tamar Herman; Tamara Best; Tara Fowler; Thomas Georgiou; Thomas Scialom; Tianhe Li; Todor Mihaylov; Tong Xiao; Ujjwal Karn; Vedanuj Goswami; Vibhor Gupta; Vignesh Ramanathan; Viktor Kerkez; Vinay Satish Kumar; Vincent Gonguet; Vish Vogeti; Vlad Poenaru; Vlad Tiberiu Mihailescu; Vladan Petrovic; Vladimir Ivanov; Wei Li; Weiwei Chu; Wenhan Xiong; Wenyin Fu; Wes Bouaziz; Whitney Meers; Will Constable; Xavier Martinet; Xiaojian Wu; Xinbo Gao; Xinfeng Xie; Xuchao Jia; Yaelle Goldschlag; Yann LeCun; Yashesh Gaur; Yasmine Babaei; Ye Qi; Yenda Li; Yi Wen; Yiwen Song; Youngjin Nam; Yuchen Hao; Yuchen Zhang; Yun Wang; Yuning Mao; Yuzi He; Zacharie Delpierre Coudert; Zachary DeVito; Zahra Hankir; Zhaoduo Wen; Zheng Yan; Zhengxing Chen; Zhenyu Yang; Zoe Papakipos
ID
Model Type ID
Text To Text
Input Type
text
Output Type
text
Description
The Llama 3 instruction tuned models are optimized for dialogue use cases and outperform many of the available open source chat llm on common industry benchmarks