unsloth/mistral-7b-instruct-v0.3
TEXT GENERATIONConcurrency Cost:1Model Size:7BQuant:FP8Ctx Length:4kPublished:May 22, 2024License:apache-2.0Architecture:Transformer0.0K Open Weights Warm

unsloth/mistral-7b-instruct-v0.3 is a 7 billion parameter instruction-tuned Mistral model, developed by Unsloth, specifically optimized for efficient fine-tuning. It leverages Unsloth's techniques to achieve 2.2x faster fine-tuning with 62% less memory usage compared to standard methods. This model is ideal for developers seeking to quickly and cost-effectively adapt a powerful Mistral-7B base for various instruction-following tasks.

Loading preview...