#codellama70b — Public Fediverse posts

Introducing Meta's Code Llama 70B, the game-changing AI programming tool!

🚀 Free for research and commercial use, will it revolutionize coding forever?

Join the revolution with Meta's #CodeLlama70B and get ready to leap #OneStepAhead into the future of AI! 🤩👨‍💻

See more here: https://twitter.com/ZNConsulting/status/1755518104027713960

#PorSiTeLoPerdiste Meta lanza CodeLlama-70B: “uno de los modelos abiertos de mayor rendimiento disponibles en la actualidad” https://www.enter.co/especiales/dev/meta-lanza-codellama-70b-uno-de-los-modelos-abiertos-de-mayor-rendimiento-disponibles-en-la-actualidad/?utm_source=dlvr.it&utm_medium=mastodon #AIDev #CODELLAMA #CodeLlama70b

Will try it locally on Apple Silicon M3 Max with different quantization method. Will post my models on #HuggingFace in a few minutes. #GGUF #codellama70b

#codellama70b now available! #metaAI

https://ai.meta.com/resources/models-and-libraries/llama-downloads/

Say you want to create #CodeLlama70b at home. #Meta says, it took 1.4 million a100-80gb #gpu hours to train. That's running 1,000 of them for close to 2 months. They're currently retailing for €18,000 each. Let's say you get a 15% discount if you buy a thousand. That'd be spending €15.3 million for the gpus alone. They're consuming 250 watts each. Running them for that many hours would rack up an electricity bill of ~€110,000. Again, that's all w/out taking the infra into account. It's crazy!

Say you want to create #CodeLlama70b at home. #Meta says, it took 1.4 million a100-80gb #gpu hours to train. That's running 1,000 of them for close to 2 months. They're currently retailing for €18,000 each. Let's say you get a 15% discount if you buy a thousand. That'd be spending €15.3 million for the gpus alone. They're consuming 250 watts each. Running them for that many hours would rack up an electricity bill of ~€110,000. Again, that's all w/out taking the infra into account. It's crazy!

#Meta releases ‘Code #Llama 70B’, an open-source behemoth to rival private AI development
#CodeLlama70B is one of the largest #opensource #AI models for code generation currently available and also happens to be the latest benchmark in the field of code generation, a branch of artificial intelligence that aims to automate the process of creating and modifying software https://venturebeat.com/ai/meta-releases-code-llama-70b-an-open-source-behemoth-to-rival-private-ai-development/

#Meta releases ‘Code #Llama 70B’, an open-source behemoth to rival private AI development
#CodeLlama70B is one of the largest #opensource #AI models for code generation currently available and also happens to be the latest benchmark in the field of code generation, a branch of artificial intelligence that aims to automate the process of creating and modifying software https://venturebeat.com/ai/meta-releases-code-llama-70b-an-open-source-behemoth-to-rival-private-ai-development/

Meta lanza CodeLlama-70B: “uno de los modelos abiertos de mayor rendimiento disponibles en la actualidad” https://www.enter.co/especiales/dev/meta-lanza-codellama-70b-uno-de-los-modelos-abiertos-de-mayor-rendimiento-disponibles-en-la-actualidad/?utm_source=dlvr.it&utm_medium=mastodon #AIDev #CODELLAMA #CodeLlama70b