What are the hardware requirements for LLaMA2 13B Estopia?

LLaMA2 13B Estopia requires approximately 26.1 GB of VRAM and supports a context window of 4K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed LLaMA2 13B Estopia and how large is it?

LLaMA2 13B Estopia is developed by KoboldAI, a model with 13b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does LLaMA2 13B Estopia perform on standard benchmarks?

LLaMA2 13B Estopia has the following published scores: MMLU 55.12. Compare against reference models on this page or on the LLM Explorer leaderboards.

LLaMA2 13B Estopia by KoboldAI — VRAM 26.1GB, 4K context

Name: LLaMA2 13B Estopia
Author: KoboldAI

LLaMA2 13B Estopia is an open-source language model by KoboldAI. Features: 13b LLM, VRAM: 26.1GB, Context: 4K, License: cc-by-nc-4.0, Quantized, HF Score: 57.2, LLM Explorer Score: 0.14, Arc: 62.3, HellaSwag: 82.5, MMLU: 55.1, TruthfulQA: 54.1, WinoGrande: 75.8, GSM8K: 13.4.

Base model:finetune:thebloke/l... Base model:thebloke/llama-2-13... Endpoints compatible Fp16 Llama Merge Mergekit Quantized Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/KoboldAI/LLaMA2-13B-Estopia

LLaMA2 13B Estopia Benchmarks

ARC: 62.29 vs 96.7 (so35)^-35.6%

HellaSwag: 82.51 vs 95.3 (gpt4)^-13.4%

MMLU: 55.12 vs 88.3 (so35)^-37.6%

TruthfulQA: 54.14 vs 59 (gpt4)^-8.2%

WinoGrande: 75.77 vs 87.5 (gpt4)^-13.4%

GSM8K: 13.42 vs 96.4 (so35)^-86.1%

LLME Score: 0.13736

^nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").

What is the LLM Explorer Rank (Score)

LLaMA2 13B Estopia (KoboldAI/LLaMA2-13B-Estopia)

🌟 Advertise your project 🚀

LLaMA2 13B Estopia Parameters and Internals

Model Type

dialogue, prose, story generation

Use Cases

Primary Use Cases:

guided narratives, storytelling, prose generation

Limitations:

response length may vary, occasional end of sequence token issues

Additional Notes

The model was developed through multiple merging stages, blending various models like UtopiaXL-13B and PygmalionAI/mythalion-13b.

Training Details

Methodology:

merge technique

Input Output

Input Format:

instruction-based prompts

Accepted Modalities:

text

Output Format:

text responses

Performance Tips:

Use suggested settings for optimal results. Adjust temperature and response length as needed.

LLM Name	LLaMA2 13B Estopia
Repository 🤗	https://huggingface.co/KoboldAI/LLaMA2-13B-Estopia
Base Model(s)	Llama 2 13B Fp16 TheBloke/Llama-2-13B-fp16
Model Size	13b
Required VRAM	26.1 GB
Updated	2026-04-20
Maintainer	KoboldAI
Model Type	llama
Model Files	10.0 GB: 1-of-3 9.7 GB: 2-of-3 6.4 GB: 3-of-3
Quantization Type	fp16
Model Architecture	LlamaForCausalLM
License	cc-by-nc-4.0
Context Length	4096
Model Max Length	4096
Transformers Version	4.36.1
Tokenizer Class	LlamaTokenizer
Vocabulary Size	32000
Torch Data Type	bfloat16

Quantized Models of the LLaMA2 13B Estopia

Model	Likes	Downloads	VRAM
LLaMA2 13B Estopia GGUF	8	207	4 GB
LLaMA2 13B Estopia GPTQ	8	11	7 GB
LLaMA2 13B Estopia AWQ	2	14	7 GB

Best Alternatives to LLaMA2 13B Estopia

Best Alternatives	Context / RAM	Downloads	Likes
Llama13b 32K Illumeet Finetune	32K / 26 GB	5	0
...Maid V3 13B 32K 6.0bpw H6 EXL2	32K / 10 GB	4	1
...Maid V3 13B 32K 8.0bpw H8 EXL2	32K / 13.2 GB	3	1
WhiteRabbitNeo 13B V1	16K / 26 GB	2921	450
CodeLlama 13B Python Fp16	16K / 26 GB	1447	25
CodeLlama 13B Instruct Fp16	16K / 26 GB	903	28
CodeLlama 13B Fp16	16K / 26 GB	38	67
...Llama 13B Instruct Hf 4bit MLX	16K / 7.8 GB	346	3
Codellama 13B Bnb 4bit	16K / 7.2 GB	88	5
Airophin 13B Pntk 16K Fp16	16K / 26 GB	380	4

Note: green Score (e.g. "73.2") means that the model is better than KoboldAI/LLaMA2-13B-Estopia.

Rank the LLaMA2 13B Estopia Capabilities

🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation

What open-source LLMs or SLMs are you in search of? 53205 in total.

Email us: info@extractum.io. Our Privacy Policy | Terms and Conditions | Suggest an improvement.

Our Social Media →

Original data from HuggingFace, OpenCompass and various public git repos.

Check out Ag3ntum — our secure, self-hosted AI agent for server management.

Release v20260328a

Support LLM Explorer

LLaMA2 13B Estopia by KoboldAI