r3m3ks/my_digital_twin
TEXT GENERATIONPricing:Input $0.37 / Cached $0.074 / Output $0.38Concurrent Unit Cost:1Model Size:8BQuant:FP8Context Size:8kTool Calling:SupportedPublished:Sep 1, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold
r3m3ks/my_digital_twin is an 8 billion parameter Llama-3 instruction-tuned causal language model developed by r3m3ks. This model was finetuned using Unsloth and Huggingface's TRL library, resulting in a 2x faster training process. It is optimized for efficient deployment and performance, leveraging its Llama-3 architecture for general language understanding and generation tasks.
Loading preview...
Model Overview
r3m3ks/my_digital_twin is an 8 billion parameter Llama-3 instruction-tuned model developed by r3m3ks. It was finetuned from unsloth/llama-3-8b-instruct-bnb-4bit using the Unsloth library in conjunction with Huggingface's TRL library. This approach enabled a significantly faster training process, specifically achieving 2x speed improvements.
Key Characteristics
- Architecture: Based on the Llama-3 family, providing robust language capabilities.
- Parameter Count: 8 billion parameters, balancing performance with computational efficiency.
- Training Efficiency: Utilizes Unsloth for accelerated finetuning, making it a good choice for developers prioritizing quick iteration and deployment.
Use Cases
- General Instruction Following: Capable of understanding and responding to a wide range of prompts due to its instruction-tuned nature.
- Efficient Deployment: Suitable for applications where faster training and potentially lower resource consumption during finetuning are beneficial.
- Llama-3 Ecosystem Integration: Benefits from the broad support and community around the Llama-3 model family.