diff --git a/unsloth/models/llama.py b/unsloth/models/llama.py index 590465ae88..de9666bca5 100644 --- a/unsloth/models/llama.py +++ b/unsloth/models/llama.py @@ -609,7 +609,7 @@ class FastLlamaModel: @staticmethod def from_pretrained( - model_name = "meta-llama/Llama-2-7b-hf", + model_name = "unsloth/llama-2-7b-bnb-4bit", max_seq_length = 4096, dtype = None, load_in_4bit = True, diff --git a/unsloth/models/loader.py b/unsloth/models/loader.py index 0ace3f494d..d458626386 100644 --- a/unsloth/models/loader.py +++ b/unsloth/models/loader.py @@ -17,10 +17,10 @@ from .mistral import FastMistralModel from transformers import AutoConfig -class FastLanguageModel: +class FastLanguageModel(FastLlamaModel): @staticmethod def from_pretrained( - model_name = "mistralai/Mistral-7B-v0.1", + model_name = "unsloth/mistral-7b-bnb-4bit", max_seq_length = 4096, dtype = None, load_in_4bit = True, diff --git a/unsloth/models/mistral.py b/unsloth/models/mistral.py index 865493edce..826ec47789 100644 --- a/unsloth/models/mistral.py +++ b/unsloth/models/mistral.py @@ -250,7 +250,7 @@ class FastMistralModel(FastLlamaModel): @staticmethod def from_pretrained( - model_name = "mistralai/Mistral-7B-v0.1", + model_name = "unsloth/mistral-7b-bnb-4bit", max_seq_length = 4096, dtype = None, load_in_4bit = True,