LuminRP-7B-128k-v0.5

LuminRP-7B-128k-v0.5

Description

LuminRP-7B-128k-v0.5 is a merge of various models that specializes in roleplaying. This model is an upgrade to Ppoyaa/LuminRP-7B-128k-v0.4 to fix the slight issue and improve the ERP and overall performance.

---

[!IMPORTANT] Based on my testing, this model outperforms the previous v0.4 version in both RP and ERP and would recommend this version over it.

This is the final version of the 7B variant as I'm pretty satisfied with it's performance as a 7B model.

---

Quants

By mradermacher:

• Static GGUF: mradermacher/LuminRP-7B-128k-v0.5-GGUF

---

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-7B-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:

<details><summary>See Merge Config</summary>

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

</details><br>