Exllamav2 quant (exl2 / 8.0 bpw) made with ExLlamaV2 v0.1.1

Other EXL2 quants:

Quant	Model Size	lm_head
2.2	3588 MB	6
2.5	3990 MB	6
3.0	4718 MB	6
3.5	5443 MB	6
3.75	5809 MB	6
4.0	6166 MB	6
4.25	6537 MB	6
5.0	7625 MB	6
6.0	9111 MB	8
6.5	9831 MB	8
8.0	11277 MB	8

LuminRP-13B-128k-v0.5

LuminRP-13B-128k-v0.5 is the 13B parameter version of the v0.5, LuminRP-7B model which specializes in RP/ERP by merging a couple models that excels in it.

Link to Ppoyaa/LuminRP-7B-128k-v0.5

This model can and will output X-rated content.

SillyTavern

Template: Alpaca, ChatML, and Mistral should be okay.

Instruct Mode: On

💻 Usage

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "Ppoyaa/LuminRP-13B-128k-v0.5"
messages = [{"role": "user", "content": "What is a large language model?"}]

tokenizer = AutoTokenizer.from_pretrained(model)
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
pipeline = transformers.pipeline(
    "text-generation",
    model=model,
    torch_dtype=torch.float16,
    device_map="auto",
)

outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
print(outputs[0]["generated_text"])

Merge Details Below:

See Merge Config

ChaoticNeutrals/BuRP_7B
Endevor/InfinityRP-v1-7B
Nitral-AI/Kunocchini-7b-128k-test
core-3/kuno-royale-v2-7b
KatyTheCutie/LemonadeRP-4.5.3
grimjim/kukulemon-7B
MaziyarPanahi/Calme-7B-Instruct-v0.9
icefog72/WestIceLemonTeaRP-32k-7b
crestf411/daybreak-kunoichi-2dpo-7b
Undi95/Mistral-RP-0.1-7B