DI

Dirty-Alice

DirtyGGUF by mradermacherUpdated 2024-08-02
Download on Hugging Face
Parameters464M
Context4K
Layers24
Architecturellama
Smallest quant0.2 GB
Quick answer

Dirty-Alice needs about 1.6 GB of memory at Q4_K_M (a 0.3 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 21 quants, from 0.2 GB (IQ1_S) to 0.4 GB (Q6_K).

Dirty‑Alice 0.464B Llama model for chat and role‑play

Dirty‑Alice is a 0.464B llama model trained on open assistant datasets and erotic role‑play text. It uses the zephyr chat format and is released under the MIT license. The model supports a 4096 token context window.

Good at

  • Chatting in a playful tone
  • Role‑play with female protagonist
  • Short‑form dialogue

Limits

  • No vision support
  • Requires 1.5–1.7 GB VRAM at 8k context
  • No mixed‑expert architecture

Best for

  • Lightweight chatbots
  • Personal role‑play assistants
  • Low‑budget local inference

Who it’s for. Developers and hobbyists who need a small, chat‑ready Llama model with a playful persona. The model is suitable for local deployment on machines with limited VRAM and for projects that require a permissive MIT license.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ1_S0.2 GB 1.1 GB
Fits Get file →
IQ1_M0.2 GB 1.1 GB
Fits Get file →
IQ2_XXS0.2 GB 1.1 GB
Fits Get file →
IQ2_XS0.2 GB 1.1 GB
Fits Get file →
IQ2_S0.2 GB 1.1 GB
Fits Get file →
IQ2_M0.2 GB 1.1 GB
Fits Get file →
IQ3_XXS0.2 GB 1.1 GB
Fits Get file →
Q2_K0.2 GB 1.2 GB
Fits Get file →
IQ3_XS0.3 GB 1.2 GB
Fits Get file →
IQ3_S0.3 GB 1.2 GB
Fits Get file →
Q3_K_S0.3 GB 1.2 GB
Fits Get file →
IQ3_M0.3 GB 1.2 GB
Fits Get file →
Q3_K_M0.3 GB 1.2 GB
Fits Get file →
Q3_K_L0.3 GB 1.2 GB
Fits Get file →
IQ4_XS0.3 GB 1.2 GB
Fits Get file →
Q4_00.3 GB 1.2 GB
Fits Get file →
Q4_K_S0.3 GB 1.2 GB
Fits Get file →
Q4_K_M0.3 GB 1.2 GB
Fits Get file →
Q5_K_S0.3 GB 1.2 GB
Fits Get file →
Q5_K_M0.4 GB 1.3 GB
Fits Get file →
Q6_K0.4 GB 1.3 GB
Fits Get file →

For 8 GB VRAM, use Q4_K_M (0.32 GB file, 1.6 GB memory) or Q3_K_M (0.28 GB file, 1.6 GB memory). For 4 GB VRAM, use IQ1_S (0.18 GB file, 1.5 GB memory) or IQ1_M (0.19 GB file, 1.5 GB memory). Larger quants like Q5_K_M or Q6_K exceed the memory budget.

What GPU do you need for Dirty-Alice?

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 (0.4 GB file, about 1.7 GB in use).

Which Dirty-Alice quant should you choose?

Q4_K_M is the usual starting point for Dirty-Alice: it balances file size and quality well. Here it is a 0.3 GB file.

If memory is tight, the smallest file is IQ1_S at 0.2 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturellama
Parameters464M
Layers24
Trained context4,096 tokens
KV cache (fp16)96 KB per token

Source

Quantized bymradermacher
FamilyDirty
Licensemit
Downloads (HF)492
Last updated2024-08-02

Frequently asked questions about Dirty-Alice

How much VRAM does Dirty-Alice need?

About 1.6 GB at Q4_K_M with an 8K context window. That is the 0.3 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 Dirty-Alice on an 8 GB GPU?

Yes, with Q6_K: the 0.4 GB file needs about 1.7 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of Dirty-Alice GGUF?

It depends on the quant: IQ1_S is 0.2 GB, IQ1_M is 0.2 GB, IQ2_XXS is 0.2 GB, IQ2_XS is 0.2 GB, IQ2_S is 0.2 GB, IQ2_M is 0.2 GB and 15 more.

What context length does Dirty-Alice support?

The GGUF header lists a trained context length of 4,096 tokens (4K). The KV cache grows with the context you use: about 0.8 GB at 8K and 0.4 GB at 4K.

Where can I download Dirty-Alice GGUF?

The files are on Hugging Face in the mradermacher/Dirty-Alice-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 Dirty-Alice use?

Hugging Face lists the license as mit. 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 Dirty‑Alice on 8 GB VRAM?

Yes. Use Q4_K_M or Q3_K_M, which require about 1.6 GB VRAM at 8k context.

Does Dirty‑Alice support vision?

No. The model has no vision capabilities.

Other GGUFs of Dirty-Alice

Similar models

Browse all →

More from mradermacher

All by mradermacher →

More Dirty models

All Dirty →

Other tiny models

All tiny models →