The LLaVa-8x7B Large Language Model (LLM) is a pretrained generative Sparse Mixture of Experts. The LLaVa-8x7B outperforms Llama 3 70B on most benchmarks we tested.
This repo contains weights that are compatible with
vLLM serving of the model as well as Hugging Face
transformers library. It is based on the original LLaVa
torrent release, but the file format and parameter names are different. Please note that model cannot (yet) be instantiated with HF.
1from transformers import AutoModelForCausalLM, AutoTokenizer
2
3model_id = "Satyam-Singh/LLaVa-Large-Language-Virtual-Assistant-v2.1.6"
4tokenizer = AutoTokenizer.from_pretrained(model_id)
5
6model = AutoModelForCausalLM.from_pretrained(model_id)
7
8text = "Hello my name is"
9inputs = tokenizer(text, return_tensors="pt")
10
11outputs = model.generate(**inputs, max_new_tokens=20)
12print(tokenizer.decode(outputs[0], skip_special_tokens=True))
By default, transformers will load the model in full precision. Therefore you might be interested to further reduce down the memory requirements to run the model through the optimizations we offer in HF ecosystem:
LLaVa-Large-Language-Virtual-Assistant-v2.1.6 is a pretrained base model and therefore does not have any moderation mechanisms.
