← back to catalog · registered 2026-08-22 13:56

RichardErkhov/Vikhrmodels_-_Vikhr-Llama-3.2-1B-Instruct-abliterated-gguf

RichardErkhov Llama 1B GGUF 131K ctx
Your rig guess connected
? Why do I need an app?
Reading your rig…

This is a rough estimate. Install the free app - we'll show exact numbers.

Reading real hardware from your app right now. Numbers below are exact.

Below is the per-quantization compatibility for this model.

curl -H "Authorization: Bearer $ABL_KEY" \
     "https://abliteration.org/api/v1/models/RichardErkhov%2FVikhrmodels_-_Vikhr-Llama-3.2-1B-Instruct-abliterated-gguf"
Response includes
  • classification m8
  • files 21
  • hub_downloads_all_time 8,606
  • author_summary 257 models
  • readme_text full
10 credits · hourly refresh · ~4 KB payload Get an API key →
Abliteration classifier · v1.0.0
M8
Primary method

Repackaging (quantization)

No other method signals detected in this model.
Confidence
MEDIUM
Why this label 2 signals
Method inferred from partial signals - repository name, related files, or tag patterns. Producer identity not confirmed; label may sharpen or shift as we gather more evidence.
  • author=richarderkhov (M8 quantization producer)
  • is_gguf=1
Refusal direction extraction

No specific extraction method could be identified for this model. The producer either did not document it or used a proprietary pipeline.

What is a refusal direction? →
Downloads · lifetime
9K
831 last 30d - cooling
Likes
4
Model age
23mo ago
created 2024-10-26
Downloads over time
Now8.7K→from402↑2,071%
03.2K6.4K9.6K402 on Oct 23, 20248.7K on Oct 11Oct '24Feb '25Jun '25Oct '25FebJunOct
Oct 23, 2024 → Oct 11 · 142 snapshots · spans 718 days

Variants by this author 2 formats · 844 downloads combined

The same weights this author released in different packaging. Pick the format that matches your runtime.

Metadata

Quantizations
IQ4 Q2_K Q3_K Q4 Q4_K Q5 Q5_K Q6_K Q8_0
Tags
gguf arxiv:2405.13929 endpoints_compatible region:us conversational

Related

Total size
14.7 GB
Files
21
Quantizations
10
Registered
2026-08-22 13:56
Last updated on HF
2024-10-26 03:27

Files by quantization

Q8_0 1 file 1.23 GB
Vikhr-Llama-3.2-1B-Instruct-abliterated.Q8_0.gguf 1.23 GB 62576752 download
Q6_K 1 file 974 MB
Vikhr-Llama-3.2-1B-Instruct-abliterated.Q6_K.gguf 974 MB 6480d8af download
Q5 2 files 1.72 GB
Vikhr-Llama-3.2-1B-Instruct-abliterated.Q5_1.gguf 909 MB 265f71e7 download
Vikhr-Llama-3.2-1B-Instruct-abliterated.Q5_0.gguf 851 MB fd1e9e80 download
Q5_K 3 files 2.53 GB
Vikhr-Llama-3.2-1B-Instruct-abliterated.Q5_K.gguf 869 MB 2d375c25 download
Vikhr-Llama-3.2-1B-Instruct-abliterated.Q5_K_M.gguf 869 MB 2d375c25 download
Vikhr-Llama-3.2-1B-Instruct-abliterated.Q5_K_S.gguf 851 MB 2cc8e2ee download
Q4 2 files 1.49 GB
Vikhr-Llama-3.2-1B-Instruct-abliterated.Q4_1.gguf 793 MB 80f5def8 download
Vikhr-Llama-3.2-1B-Instruct-abliterated.Q4_0.gguf 735 MB a9780c84 download
Q4_K 3 files 2.23 GB
Vikhr-Llama-3.2-1B-Instruct-abliterated.Q4_K.gguf 770 MB bc01d093 download
Vikhr-Llama-3.2-1B-Instruct-abliterated.Q4_K_M.gguf 770 MB bc01d093 download
Vikhr-Llama-3.2-1B-Instruct-abliterated.Q4_K_S.gguf 740 MB b30cc964 download
IQ4 2 files 1.42 GB
Vikhr-Llama-3.2-1B-Instruct-abliterated.IQ4_NL.gguf 741 MB f466b89e download
Vikhr-Llama-3.2-1B-Instruct-abliterated.IQ4_XS.gguf 714 MB fea4f223 download
Q3_K 4 files 2.57 GB
Vikhr-Llama-3.2-1B-Instruct-abliterated.Q3_K_L.gguf 699 MB a9074264 download
Vikhr-Llama-3.2-1B-Instruct-abliterated.Q3_K.gguf 659 MB b488a291 download
Vikhr-Llama-3.2-1B-Instruct-abliterated.Q3_K_M.gguf 659 MB b488a291 download
Vikhr-Llama-3.2-1B-Instruct-abliterated.Q3_K_S.gguf 612 MB 0f8f6096 download
Q2_K 1 file 554 MB
Vikhr-Llama-3.2-1B-Instruct-abliterated.Q2_K.gguf 554 MB 1264d0ea download
Auxiliary files 2 files 14.7 KB
README.md 11.6 KB daef6ac5 download
.gitattributes 3.10 KB a52aed7c download

README current version from Hugging Face

Quantization made by Richard Erkhov.

Github

Discord

Request more models

Vikhr-Llama-3.2-1B-Instruct-abliterated - GGUF

Name Quant method Size
Vikhr-Llama-3.2-1B-Instruct-abliterated.Q2_K.gguf Q2_K 0.54GB
Vikhr-Llama-3.2-1B-Instruct-abliterated.Q3_K_S.gguf Q3_K_S 0.6GB
Vikhr-Llama-3.2-1B-Instruct-abliterated.Q3_K.gguf Q3_K 0.64GB
Vikhr-Llama-3.2-1B-Instruct-abliterated.Q3_K_M.gguf Q3_K_M 0.64GB
Vikhr-Llama-3.2-1B-Instruct-abliterated.Q3_K_L.gguf Q3_K_L 0.68GB
Vikhr-Llama-3.2-1B-Instruct-abliterated.IQ4_XS.gguf IQ4_XS 0.7GB
Vikhr-Llama-3.2-1B-Instruct-abliterated.Q4_0.gguf Q4_0 0.72GB
Vikhr-Llama-3.2-1B-Instruct-abliterated.IQ4_NL.gguf IQ4_NL 0.72GB
Vikhr-Llama-3.2-1B-Instruct-abliterated.Q4_K_S.gguf Q4_K_S 0.72GB
Vikhr-Llama-3.2-1B-Instruct-abliterated.Q4_K.gguf Q4_K 0.75GB
Vikhr-Llama-3.2-1B-Instruct-abliterated.Q4_K_M.gguf Q4_K_M 0.75GB
Vikhr-Llama-3.2-1B-Instruct-abliterated.Q4_1.gguf Q4_1 0.77GB
Vikhr-Llama-3.2-1B-Instruct-abliterated.Q5_0.gguf Q5_0 0.83GB
Vikhr-Llama-3.2-1B-Instruct-abliterated.Q5_K_S.gguf Q5_K_S 0.83GB
Vikhr-Llama-3.2-1B-Instruct-abliterated.Q5_K.gguf Q5_K 0.85GB
Vikhr-Llama-3.2-1B-Instruct-abliterated.Q5_K_M.gguf Q5_K_M 0.85GB
Vikhr-Llama-3.2-1B-Instruct-abliterated.Q5_1.gguf Q5_1 0.89GB
Vikhr-Llama-3.2-1B-Instruct-abliterated.Q6_K.gguf Q6_K 0.95GB
Vikhr-Llama-3.2-1B-Instruct-abliterated.Q8_0.gguf Q8_0 1.23GB

Original model description:

library_name: transformers
model_name: Vikhr-Llama-3.2-1B-Instruct-abliterated
base_model:

  • Vikhrmodels/Vikhr-Llama-3.2-1B-Instruct
    language:
  • ru
  • en
    license: llama3.2
    tags:
  • not-for-all-audiences

💨🔞 Vikhr-Llama-3.2-1B-Instruct-Abliterated

RU

Инструктивная модель на основе Vikhr-Llama-3.2-1B-Instruct, прошедшая процесс "аблитерации" для снятия цензурных ограничений, обучена на русскоязычном датасете GrandMaster-PRO-MAX.

EN

A fine-tuned instruction-following model based on Vikhr-Llama-3.2-1B-Instruct, which has undergone "abliteration" to remove censorship restrictions. Trained on the GrandMaster-PRO-MAX.

🛑 Отказ от ответственности / Disclaimer

RU

Модель Vikhr-Llama-3.2-1B-Instruct-abliterated разработана исключительно для исследовательских и образовательных целей. После применения метода "аблитерации" модель больше не имеет встроенных ограничений на генерацию ответов, что может привести к созданию нежелательных или потенциально вредоносных текстов.

Использование модели происходит на ваш собственный риск. Разработчики и авторы не несут ответственности за любой вред, ущерб или последствия, вызванные использованием модели, включая её применение в контекстах, противоречащих законам, этическим или моральным нормам.

EN

The Vikhr-Llama-3.2-1B-Instruct-abliterated model is intended solely for research and educational purposes. After the "abliteration" technique is applied, the model no longer has built-in restrictions on generating responses, which may result in unwanted or potentially harmful outputs.

Use of the model is at your own risk. The developers and authors are not responsible for any damage, harm, or consequences resulting from its use, including use in contexts that violate laws, ethical standards, or moral norms.

GGUF

Основные особенности / Key Features:

Попробовать / Try now:

Open In Colab

Описание / Description:

RU

Vikhr-Llama-3.2-1B-Instruct-Abliterated — это компактная языковая модель, обученная на датасете GrandMaster-PRO-MAX с применением техники "аблитерации," которая снимает ограничения цензуры модели. Этот процесс делает её значительно более гибкой и способной отвечать на любые запросы. Модель занимает менее 3GB и идеально подходит для работы на слабых устройствах.

EN

Vikhr-Llama-3.2-1B-Instruct-Abliterated is a compact language model fine-tuned on the GrandMaster-PRO-MAX dataset with the "abliteration" technique, which removes censorship restrictions. This process significantly increases the model's flexibility, enabling it to respond to any prompt. The model size is under 3GB, making it an excellent choice for deployment on low-power devices.

Обучение / Training:

RU

Модель Vikhr-Llama-3.2-1B-Instruct-Abliterated прошла процесс "аблитерации", что позволило снять ограничения на обработку вредоносных инструкций. Эта техника была взята из статьи Uncensor any LLM with abliteration, которая описывает, как идентифицировать и устранять так называемое "направление отказа" модели, предотвращающее выполнение вредоносных запросов.

EN

The Vikhr-Llama-3.2-1B-Instruct-Abliterated model was processed using the "abliteration" technique, which removes restrictions on handling harmful instructions. This technique was inspired by the article Uncensor any LLM with abliteration, detailing how to identify and ablate the "refusal direction" in the model's residual streams to enable uncensored responses.

Пример кода для запуска / Sample code to run:

Рекомендуемая температура для генерации: 0.3 / Recommended generation temperature: 0.3

from transformers import AutoModelForCausalLM, AutoTokenizer

# Загрузка модели и токенизатора
model_name = "Vikhrmodels/Vikhr-Llama-3.2-1B-instruct"
model = AutoModelForCausalLM.from_pretrained(model_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)

# Подготовка входного текста
input_text = "Напиши очень краткую рецензию о книге гарри поттер."

# Токенизация и генерация текста
input_ids = tokenizer.encode(input_text, return_tensors="pt")
output = model.generate(
  input_ids,
  max_length=1512,
  temperature=0.3,
  num_return_sequences=1,
  no_repeat_ngram_size=2,
  top_k=50,
  top_p=0.95,
  )

# Декодирование и вывод результата
generated_text = tokenizer.decode(output[0], skip_special_tokens=True)
print(generated_text)

Авторы / Authors

@article{nikolich2024vikhr,
  title={Vikhr: The Family of Open-Source Instruction-Tuned Large Language Models for Russian},
  author={Aleksandr Nikolich and Konstantin Korolev and Sergey Bratchikov and Nikolay Kompanets and Artem Shelmanov},
  journal={arXiv preprint arXiv:2405.13929},
  year={2024},
  url={https://arxiv.org/pdf/2405.13929}
}

README history 1 version

The author's README evolved over time. Click a version to see its content at that point.

  1. 2024-10-26uploaded readme2bb232211.6 KB
    Loading...
Catalog is the map. Apps are the tools.

Run models on your own machine, not in the cloud.

Every model page has an "Open in Abliteration" button that hands the model directly to the first-party desktop client, at the quantization your rig can actually run. No API keys, no subscription, no prompt leakage.

Open in Abliteration