Llama-3.3-70B-Instruct is a large language model released by Meta, released under the Llama 3.3 community license (gated) license, with a claimed knowledge cutoff of 2024-12-06. Its weights, configuration, and tokenizer files are shortlisted for hashing and on-chain attestation by Weight Registry.
Description
Meta's 70B instruction-tuned model, notable for matching much larger Llama 3.1 405B performance on many benchmarks.
Metadata
Hugging Face stats
921,237 downloads on Hugging Face · 3,061 likes · pipeline: text-generation · library: transformers
Weight Hashes
Not yet hashed -- queued for processing.