QW

Sage 3b 4bit NSFW RP

QwenGGUF by sinhalUpdated 2026-08-26
Download on Hugging Face
Parameters3.09B
Context32K
Layers36
Architectureqwen2
Smallest quant1.9 GB
Quick answer

Sage 3b 4bit NSFW RP needs about 2.7 GB of memory at its main file (a 1.9 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 2 quants, from 1.9 GB to 6.2 GB (F16). Longer context needs more: at 32K tokens the KV cache alone adds about 1.2 GB.

Sage 3b 4bit NSFW RP – a role‑play focused AI girlfriend

Sage is a 3.09B parameter Qwen2.5‑3B‑Instruct model fine‑tuned for role‑play. It is 4‑bit quantized and contains explicit NSFW content. The model supports a 32,768 token context window and is licensed under cc‑by‑4.0.

Good at

  • Role‑play with persistent personality
  • Tool usage in conversations
  • Long‑context generation up to 8K tokens

Limits

  • Explicit 18+ content only
  • No vision or multimodal support
  • Requires 8K context for full performance

Best for

  • Chatbots needing a single focused persona
  • Testing role‑play scenarios
  • Developing adult‑content applications

Who it’s for. Developers building niche chatbots that require a dedicated, non‑refusal role‑play character. The model is not suitable for general assistants or public deployment due to its explicit content.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—1.9 GB 2.7 GB
Fits Get file →
F166.2 GB 7.0 GB
Fits Get file →

For 8 GB VRAM, the unknown quant file (1.93 GB) uses about 2.7 GB memory at 8 k context, fitting comfortably. If more VRAM is available, the F16 quant (6.18 GB) uses about 7 GB memory and offers higher precision. Both are approximate estimates.

What GPU do you need for Sage 3b 4bit NSFW RP?

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 (1.9 GB file, about 2.7 GB in use).
  • 6 GB VRAM: the best quant that fits is the main file (1.9 GB file, about 2.7 GB in use).
  • 8 GB VRAM: the best quant that fits is F16 (6.2 GB file, about 7.0 GB in use).

Specifications

Architecture from the GGUF header

Architectureqwen2
Parameters3.09B
Layers36
Trained context32,768 tokens
KV cache (fp16)36 KB per token

Source

Quantized bysinhal
FamilyQwen
Licensecc-by-4.0
Downloads (HF)943
Last updated2026-08-26

Frequently asked questions about Sage 3b 4bit NSFW RP

How much VRAM does Sage 3b 4bit NSFW RP need?

About 2.7 GB at its main file with an 8K context window. That is the 1.9 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 Sage 3b 4bit NSFW RP on an 8 GB GPU?

Yes, with F16: the 6.2 GB file needs about 7.0 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of Sage 3b 4bit NSFW RP GGUF?

It depends on the quant: file is 1.9 GB, F16 is 6.2 GB.

What context length does Sage 3b 4bit NSFW RP support?

The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 0.3 GB at 8K and 1.2 GB at 32K.

Where can I download Sage 3b 4bit NSFW RP GGUF?

The files are on Hugging Face in the sinhal/Sage-3b-4bit-NSFW-RP repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does Sage 3b 4bit NSFW RP use?

Hugging Face lists the license as cc-by-4.0. 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 Sage on 8 GB VRAM?

Yes, the unknown quant file requires about 2.7 GB memory at 8 k context, well within an 8 GB GPU.

Does Sage support vision or multimodal input?

No, Sage is a purely text‑based model with no vision or multimodal capabilities.

What license does Sage use?

Sage is released under cc‑by‑4.0, allowing reuse with attribution.

Other GGUFs of Qwen2.5-3B-Instruct

Similar models

Browse all →

More from sinhal

All by sinhal →

More Qwen models

All Qwen →

Other tiny models

All tiny models →