680x514px 223.96KB 1000x1000px 940.28KB 496x842px 266.14KB 892x1452px 417.95KB 1000x1000px 39.77KB 5400x3600px 453.63KB 1979x2086px 309.7KB 669x1600px 130.46KB 2866x3146px 7.19MB 894x894px 389.88KB 860x1078px 1.01MB 647x870px 401.09KB 1977x2313px 1.47MB 570x713px 501.61KB 640x400px 74.26KB 500x588px 222.28KB 509x583px 401.56KB 999x1500px 501.16KB 500x580px 559.59KB 839x946px 57.39KB 822x636px 893.48KB 938x1019px 138.19KB 695x1080px 51.59KB 461x627px 210.76KB 557x498px 13.88KB 512x512px 95.42KB 500x664px 208.01KB 526x635px 404.01KB 517x600px 334.13KB 420x598px 41.72KB 1141x1280px 62.37KB 502x519px 193.73KB 552x602px 202.92KB 833x1280px 227.24KB 747x1024px 473.31KB 512x512px 109KB 500x580px 252.01KB 705x903px 389.23KB 515x689px 162.34KB 2636x1660px 393.14KB 500x750px 200.77KB 499x648px 184.72KB 1161x1280px 325.62KB 500x774px 220.97KB 712x988px 289.01KB 768x600px 289.11KB 512x512px 122.2KB 1031x2400px 288.85KB 600x522px 18.3KB 540x720px 328.4KB 4757x5000px 1014.56KB 894x894px 131.77KB 541x743px 620.8KB 512x512px 69.24KB 640x400px 74.26KB 673x1024px 176.51KB 416x419px 67.45KB 1969x2076px 307.93KB 1080x810px 303.5KB 880x880px 44.66KB
Image of a llama face
Introducing Llama 2
The next generation of our
open source large language model
Llama 2 is available for free for research and commercial use.
Inside the model
This release includes model weights and starting code for pretrained and fine-tuned Llama language models (Llama Chat, Code Llama) — ranging from 7B to 70B parameters.
Llama 2 pretrained models are trained on 2 trillion tokens, and have double the context length than Llama 1. Its fine-tuned models have been trained on over 1 million human annotations.
Llama 2 outperforms other open source language models on many external benchmarks, including reasoning, coding, proficiency, and knowledge tests.
More model details
Llama 2 was pretrained on publicly available online data sources.
The fine-tuned model, Llama Chat, leverages publicly available instruction datasets and over 1 million human annotations.
Code Llama is a code generation model built on Llama 2, trained on 500B tokens of code. It supports common programming languages being used today, including Python, C++, Java, PHP, Typescript (Javascript), C#, and Bash.
Partnerships
Our global partners and supporters
We have a broad range of supporters around the world who believe in our open approach to today’s AI — companies that have given early feedback and are excited to build with Llama 2, cloud providers that will include the model as part of their offering to customers, researchers committed to doing research with the model, and people across tech, academia, and policy who see the benefits of Llama and an open platform as we do.
Statement of support for Meta’s open approach to today’s AI
“We support an open innovation approach to AI. Responsible and open innovation gives us all a stake in the AI development process, bringing visibility, scrutiny and trust to these technologies. Opening today’s Llama models will let everyone benefit from this technology.”