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.
Fine-tuningDocs | Llama 3 70B Instruct can be customized with your data to improve responses. Fireworks uses LoRA to efficiently train and deploy your personalized model |
On-demand DeploymentDocs | On-demand deployments allow you to use Llama 3 70B Instruct on dedicated GPUs with Fireworks' high-performance serving stack with high reliability and no rate limits. |
Llama 3 70B Instruct is an instruction-tuned large language model developed by Meta, part of the Llama 3 family released in April 2024. It is optimized for assistant-style dialogue and natural language generation tasks.
Llama 3 70B Instruct excels at:
The model supports a context length of 8.2k tokens.
Usable context is up to 8.2k tokens, which matches the maximum context length.
The model supports 48 quantized variants, confirming availability in 4-bit and 8-bit formats.
Despite extensive safety evaluations, the model may:
No, function calling is not supported.
The model has 70.6 billion parameters.
Llama 3 70B Instruct supports LoRA fine-tuning, full fine-tuning, and serverless LoRA.
Tokens are billed per 1M input/output tokens.
The model is released under the Llama 3 Community License, which allows commercial use. License details are hosted at llama.meta.com/license.
