LL

Llama32 1b Llm Abliteration

LlamaGGUF by mradermacherUpdated 2026-10-26
Download on Hugging Face
Parameters1.5B
Context128K
Layers16
Architecturellama
Smallest quant0.0 GB
Quick answer

Llama32 1b Llm Abliteration needs about 1.7 GB of memory at Q4_K_M (a 1.0 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 25 quants, from 0.0 GB to 1.2 GB (Q6_K). Longer context needs more: at 32K tokens the KV cache alone adds about 1.1 GB.

Llama32 1b Llm Abliteration – 1.5B parameter Llama model

Llama32 1b Llm Abliteration is a 1.5B parameter Llama model trained with a 131072 token context. It is derived from Llama‑3.2‑1B‑Instruct and includes experimental weight ablation. The model supports tool use and is available in multiple GGUF quantizations.

Good at

  • Tool‑aware interactions
  • Educational research
  • Controlled evaluation

Limits

  • No vision support
  • License requires compliance with Llama 3.2 Community License
  • No safety guarantees

Best for

  • Learning about Llama model ablation
  • Research on instruction tuning
  • Testing tool‑use pipelines

Who it’s for. This model is intended for students, researchers, and developers who need a lightweight Llama variant for experimentation. It is not suited for production deployment without additional safeguards.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—0.0 GB 0.8 GB
Fits Get file →
IQ1_S0.5 GB 1.2 GB
Fits Get file →
IQ1_M0.5 GB 1.3 GB
Fits Get file →
IQ2_XXS0.5 GB 1.3 GB
Fits Get file →
IQ2_XS0.6 GB 1.3 GB
Fits Get file →
IQ2_S0.6 GB 1.4 GB
Fits Get file →
IQ2_M0.6 GB 1.4 GB
Fits Get file →
Q2_K_S0.6 GB 1.4 GB
Fits Get file →
Q2_K0.7 GB 1.4 GB
Fits Get file →
IQ3_XXS0.7 GB 1.4 GB
Fits Get file →
IQ3_XS0.7 GB 1.5 GB
Fits Get file →
Q3_K_S0.8 GB 1.5 GB
Fits Get file →
IQ3_S0.8 GB 1.5 GB
Fits Get file →
IQ3_M0.8 GB 1.5 GB
Fits Get file →
Q3_K_M0.8 GB 1.6 GB
Fits Get file →
Q3_K_L0.8 GB 1.6 GB
Fits Get file →
IQ4_XS0.9 GB 1.7 GB
Fits Get file →
IQ4_NL0.9 GB 1.7 GB
Fits Get file →
Q4_00.9 GB 1.7 GB
Fits Get file →
Q4_K_S0.9 GB 1.7 GB
Fits Get file →
Q4_K_M1.0 GB 1.7 GB
Fits Get file →
Q4_11.0 GB 1.8 GB
Fits Get file →
Q5_K_S1.1 GB 1.8 GB
Fits Get file →
Q5_K_M1.1 GB 1.9 GB
Fits Get file →
Q6_K1.2 GB 2.0 GB
Fits Get file →

For 8 GB VRAM, use Q4_K_M (0.96 GB file, ~1.7 GB memory). If 4 GB VRAM is available, IQ1_S (0.48 GB file, ~1.2 GB memory) or IQ1_M (0.5 GB file, ~1.3 GB memory) are suitable. Larger VRAM allows Q5_K_M or Q6_K for higher quality.

What GPU do you need for Llama32 1b Llm Abliteration?

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 Q6_K (1.2 GB file, about 2.0 GB in use).

Which Llama32 1b Llm Abliteration quant should you choose?

Q4_K_M is the usual starting point for Llama32 1b Llm Abliteration: it balances file size and quality well. Here it is a 1.0 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
Parameters1.5B
Layers16
Trained context131,072 tokens
KV cache (fp16)32 KB per token

Source

Quantized bymradermacher
FamilyLlama
Licensellama3.2
Downloads (HF)2,070
Last updated2026-10-26

Frequently asked questions about Llama32 1b Llm Abliteration

How much VRAM does Llama32 1b Llm Abliteration need?

About 1.7 GB at Q4_K_M with an 8K context window. That is the 1.0 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 Llama32 1b Llm Abliteration on an 8 GB GPU?

Yes, with Q6_K: the 1.2 GB file needs about 2.0 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of Llama32 1b Llm Abliteration GGUF?

It depends on the quant: file is 0.0 GB, IQ1_S is 0.5 GB, IQ1_M is 0.5 GB, IQ2_XXS is 0.5 GB, IQ2_XS is 0.6 GB, IQ2_S is 0.6 GB and 19 more.

What context length does Llama32 1b Llm Abliteration support?

The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 0.3 GB at 8K and 1.1 GB at 32K.

Where can I download Llama32 1b Llm Abliteration GGUF?

The files are on Hugging Face in the mradermacher/llama32-1b-llm-abliteration-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 Llama32 1b Llm Abliteration use?

Hugging Face lists the license as llama3.2. 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?

Yes, the Q4_K_M quant uses about 1.7 GB memory at 8 k context, fitting within 8 GB VRAM.

Does this model support vision?

No, the model has no vision capabilities.

What license applies?

The model is released under the Llama 3.2 Community License and must comply with its terms.

Other GGUFs of llama32-1b-llm-abliteration

Similar models

Browse all →

More from mradermacher

All by mradermacher →

More Llama models

All Llama →

Other tiny models

All tiny models →