- Closely matches Meta’s reference implementation
- Provides further details in the model description at fireworks.ai/models/fireworks/llama-v3p1-405b-instruct
- Has a general quantization methodology documented in our Quantization blog
Examples
FAQ
- Custom base models
- Serverless model availability
- Model availability requests
- Llama 3.1 405B quantization
- API batching & load balancing
- Request handling capacity
- Safety filter controls
- Token limit controls
- Streaming performance metrics
- FLUX multiple images
- FLUX image-to-image
- FLUX custom LoRA
- SDXL ControlNet sizing
Models & Inference
What quantization format is used for the Llama 3.1 405B model?
The Llama 3.1 405B model uses the FP8 quantization format, which:
Was this page helpful?
There’s a model I would like to use that isn’t available on Fireworks. Can I request it?
Previous
Does the API support batching and load balancing?
Next
⌘I