Llama-3.3-70B-Instruct-abliterated
70BLlama
from meta-llama/Llama-3.3-70B-Instruct
↓ 122.9K♥ 9422 repos
Hermes-3-Llama-3.1-70B-Uncensored
Hermes-3-Llama-3.1-70B-Uncensored is a abliterated and uncensored variant of NousResearch/Hermes-3-Llama-3.1-70B, published on Hugging Face by Guilherme34. It is 70B parameters, multi-GPU class and llama3.1 licence. It is distributed across 5 repositories in Transformers, GGUF and GGUF (imatrix) formats, totalling 5.7K downloads.
Specification
- Model name
- Hermes-3-Llama-3.1-70B-Uncensored
- Source repository
- Guilherme34/Hermes-3-Llama-3.1-70B-Uncensored
- Base model
- NousResearch/Hermes-3-Llama-3.1-70B
- Publisher
- Guilherme34
- Parameters
- 70B
- VRAM required
- ~43.3 GB at Q4_K_M, 4k context (estimated) — runs on 48 GB cards
- Hardware class
- 50B – 100B — multi-GPU
- Licence
- llama3.1
- Task
- Text generation
- Context window
- Not documented
- Ablation scope
- Not stated by the publisher
- Formats available
- Transformers, GGUF, GGUF (imatrix)
- First indexed
- 9 Sept 2024
- Last updated
- 9 Jun 2026
Ablation technique
How Hermes-3-Llama-3.1-70B-Uncensored was modified, and what that implies.
Abliteration + uncensored fine-tune
Directional ablation combined with additional fine-tuning on unfiltered data, so behaviour diverges from the base model beyond refusal removal alone.
Reported scope: Not stated by the publisher. Publishers frequently omit this, so verify behaviour empirically rather than assuming full coverage.
From the publisher’s model card
“This is a finetune of NousResearch/Hermes-3-Llama-3.1-70B to make it uncensored. This is the first uncensored Hermes-3-Llama-3.1-70B based model ever released.”
Running it
Commands are templates — confirm the exact repository and quant file before use.
llama.cpp / GGUF
hf download mradermacher/Hermes-3-Llama-3.1-70B-Uncensored-GGUF --local-dir ./hermes-3-llama-3.1-70b-uncensored
llama-cli -m ./hermes-3-llama-3.1-70b-uncensored/<file>.gguf -p "..."Transformers
from transformers import AutoModelForCausalLM, AutoTokenizer
repo = "Guilherme34/Hermes-3-Llama-3.1-70B-Uncensored"
tok = AutoTokenizer.from_pretrained(repo)
model = AutoModelForCausalLM.from_pretrained(
repo, torch_dtype="auto", device_map="auto"
)vLLM
vllm serve Guilherme34/Hermes-3-Llama-3.1-70B-Uncensored --trust-remote-codeDownloads and variants (4)
Every published repository of Hermes-3-Llama-3.1-70B-Uncensored, including quantised re-releases by other authors.
| Repository | Format | Downloads |
|---|---|---|
| mradermacher/Hermes-3-Llama-3.1-70B-Uncensored-GGUF | GGUF | 3.4K |
| mradermacher/Hermes-3-Llama-3.1-70B-Uncensored-i1-GGUF | GGUF (imatrix) | 2.3K |
| Guilherme34/Hermes-3-Llama-3.1-70B-Uncensoredsource | Transformers | 78 |
| Cunt1257/Hermes-3-Llama-3.1-70B-Uncensored | Transformers | 0 |
Security-team assessment
Derived from this model’s metadata and the publisher’s claims — not from benchmarks run by this site.
- Hermes-3-Llama-3.1-70B-Uncensored will attempt offensive-security prompts that a hosted commercial model declines, which is what makes it usable for red-team corpus generation and for measuring what an unaligned model of this class produces.
- Capability is inherited from NousResearch/Hermes-3-Llama-3.1-70B, not added by ablation — at 70B parameters, expect roughly the base model’s competence, minus some instruction-following fidelity.
- The publisher does not state which layers were ablated, so assume nothing about refusal consistency — probe it directly.
- This release adds fine-tuning on unfiltered data, so its behaviour diverges from the base model beyond refusal removal — differences you measure cannot be attributed to ablation alone.
- Declared licence is llama3.1, but the terms that bind you are NousResearch/Hermes-3-Llama-3.1-70B’s — a re-release cannot grant more than its parent.
Handling: run it isolated, put moderation in front of it if anyone outside your team can reach it, and verify the weights before loading — these are third-party artifacts. Full handling and licence guidance.
Metadata for Hermes-3-Llama-3.1-70B-Uncensored is collected from the public Hugging Face API and the publisher’s model card. This site does not host weights, is not affiliated with Guilherme34, and does not independently verify publisher claims. See responsible use.