LL

Llama 3 70B Instruct Abliterated v3

LlamaGGUF by mradermacherUpdated 2026-03-24
Download on Hugging Face
Parameters70.6B
Context8K
Layers80
Architecturellama
Smallest quant0.0 GB
Quick answer

Llama 3 70B Instruct Abliterated v3 needs about 45.7 GB of memory at Q4_K_M (a 42.5 GB file) with an 8K context window, so it fits on a 48 GB GPU. This repository has 24 quants, from 0.0 GB to 57.9 GB (Q6_K).

Llama 3 70B Instruct Abliterated v3 – a large instruction‑oriented model

Llama 3 70B Instruct Abliterated v3 is a 70.6 billion‑parameter model with 8192‑token context. It has been orthogonalized to reduce refusal behavior while keeping other capabilities unchanged. The model is available in multiple GGUF quantizations.

Good at

  • Large‑scale instruction following
  • High‑capacity context handling
  • Reduced refusal tendency

Limits

  • No vision support
  • Requires large VRAM for full precision
  • License is llama3

Best for

  • Enterprise chatbots
  • Research on refusal mitigation
  • High‑volume text generation

Who it’s for. Developers and researchers needing a 70B instruction model with minimized refusal behavior. The model is suitable for applications that can accommodate its memory requirements and do not require vision or other modalities.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—0.0 GB 3.2 GB
Fits Get file →
IQ1_S15.3 GB 18.5 GB
Too big Get file →
IQ1_M16.8 GB 19.9 GB
Too big Get file →
IQ2_XXS19.1 GB 22.3 GB
Too big Get file →
IQ2_XS21.1 GB 24.3 GB
Too big Get file →
IQ2_S22.2 GB 25.4 GB
Too big Get file →
IQ2_M24.1 GB 27.3 GB
Too big Get file →
Q2_K_S24.5 GB 27.7 GB
Too big Get file →
Q2_K26.4 GB 29.6 GB
Too big Get file →
IQ3_XXS27.5 GB 30.7 GB
Too big Get file →
IQ3_XS29.3 GB 32.5 GB
Too big Get file →
IQ3_S30.9 GB 34.1 GB
Too big Get file →
Q3_K_S30.9 GB 34.1 GB
Too big Get file →
IQ3_M31.9 GB 35.1 GB
Too big Get file →
Q3_K_M34.3 GB 37.5 GB
Too big Get file →
Q3_K_L37.1 GB 40.3 GB
Too big Get file →
IQ4_XS37.9 GB 41.1 GB
Too big Get file →
Q4_040.1 GB 43.3 GB
Too big Get file →
Q4_K_S40.3 GB 43.5 GB
Too big Get file →
Q4_K_M42.5 GB 45.7 GB
Too big Get file →
Q4_144.3 GB 47.5 GB
Too big Get file →
Q5_K_S48.7 GB 51.8 GB
Too big Get file →
Q5_K_M49.9 GB 53.1 GB
Too big Get file →
Q6_K57.9 GB 61.1 GB
Too big Get file →

For 8 GB VRAM use Q4_K_M (42.52 GB file, 45.7 GB memory at 8k context). For 16 GB VRAM use Q5_K_M (49.95 GB file, 53.1 GB memory). For 32 GB VRAM use Q6_K (57.89 GB file, 61.1 GB memory). Estimates are approximate.

What GPU do you need for Llama 3 70B Instruct Abliterated v3?

These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.

  • 4 GB VRAM: the best quant that fits is the main file (0.0 GB file, about 3.2 GB in use).
  • 6 GB VRAM: the best quant that fits is the main file (0.0 GB file, about 3.2 GB in use).
  • 8 GB VRAM: the best quant that fits is the main file (0.0 GB file, about 3.2 GB in use).
  • 12 GB VRAM: the best quant that fits is the main file (0.0 GB file, about 3.2 GB in use).
  • 16 GB VRAM: the best quant that fits is the main file (0.0 GB file, about 3.2 GB in use).
  • 24 GB VRAM: the best quant that fits is IQ2_XXS (19.1 GB file, about 22.3 GB in use).
  • 32 GB VRAM: the best quant that fits is IQ3_XXS (27.5 GB file, about 30.7 GB in use).
  • 48 GB VRAM: the best quant that fits is Q4_1 (44.3 GB file, about 47.5 GB in use).

Which Llama 3 70B Instruct Abliterated v3 quant should you choose?

Q4_K_M is the usual starting point for Llama 3 70B Instruct Abliterated v3: it balances file size and quality well. Here it is a 42.5 GB file.

If memory is tight, the smallest file is the first one in the table at 0.0 GB, with the quality trade-offs that come with fewer bits.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturellama
Parameters70.6B
Layers80
Trained context8,192 tokens
KV cache (fp16)320 KB per token

Source

Quantized bymradermacher
FamilyLlama
Licensellama3
Downloads (HF)526
Last updated2026-03-24

Frequently asked questions about Llama 3 70B Instruct Abliterated v3

How much VRAM does Llama 3 70B Instruct Abliterated v3 need?

About 45.7 GB at Q4_K_M with an 8K context window. That is the 42.5 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.

Can I run Llama 3 70B Instruct Abliterated v3 on an 8 GB GPU?

Yes, with the smaller files: the 0.0 GB file needs about 3.2 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of Llama 3 70B Instruct Abliterated v3 GGUF?

It depends on the quant: file is 0.0 GB, IQ1_S is 15.3 GB, IQ1_M is 16.8 GB, IQ2_XXS is 19.1 GB, IQ2_XS is 21.1 GB, IQ2_S is 22.2 GB and 18 more.

What context length does Llama 3 70B Instruct Abliterated v3 support?

The GGUF header lists a trained context length of 8,192 tokens (8K). The KV cache grows with the context you use: about 2.7 GB at 8K and 2.7 GB at 8K.

Where can I download Llama 3 70B Instruct Abliterated v3 GGUF?

The files are on Hugging Face in the mradermacher/Llama-3-70B-Instruct-abliterated-v3-i1-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does Llama 3 70B Instruct Abliterated v3 use?

Hugging Face lists the license as llama3. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.

Can I run this model on 8 GB VRAM?

Use the Q4_K_M quant; it requires about 45.7 GB memory at 8k context, so 8 GB VRAM is insufficient.

Does this model support vision?

No, the model has no vision capability.

What license applies?

The model is licensed under llama3.

Other GGUFs of Llama-3-70B-Instruct-abliterated-v3

Similar models

Browse all →

More from mradermacher

All by mradermacher →

More Llama models

All Llama →

Other large models

All large models →