WineryLabs/Winery-Qwen2.5-1.5B-RP-Soup-GGUF overview
<p align="center" <img src="winery logo.png" width="180"/ </p 🍷 Winery Qwen2.5 1.5B · RP Soup An equal parts model soup of eight Qwen2.5 1.5B roleplay, story …
Runs locally from ~1.53 GB disk (4 GB VRAM class GPUs with llama.cpp / guIDE).
Repository Files & Downloads
| File | Type | Quantization | Size | Link |
|---|---|---|---|---|
| Winery-Qwen2.5-1.5B-RP-Soup-Q8_0.gguf | GGUF | Q8_0 | 1.53 GB | Download |
Model Details
| Model ID | WineryLabs/Winery-Qwen2.5-1.5B-RP-Soup-GGUF |
|---|---|
| Author | WineryLabs |
| Pipeline | text-generation |
| License | apache-2.0 |
| Base model | EVA-UNIT-01/EVA-Qwen2.5-1.5B-v0.0,EVA-UNIT-01/EVA-D-Qwen2.5-1.5B-v0.0,ReXeeD/Luminus-1.5B-Roleplay,BeaverAI/Qwen2.5-QwQ-RP-Draft-v0.2-1.5B,dphn/Dolphin3.0-Qwen2.5-1.5B,Goekdeniz-Guelmez/Josiefied-Qwen2.5-1.5B-Instruct-abliterated-v1,nbeerbower/EVA-abliterated-TIES-Qwen2.5-1.5B,NathanRoll/writing-rlvr-qwen2.5-1.5b |
| Last modified | 2026-10-03T02:15:45.000Z |
Model README
---
license: apache-2.0
base_model:
- EVA-UNIT-01/EVA-Qwen2.5-1.5B-v0.0
- EVA-UNIT-01/EVA-D-Qwen2.5-1.5B-v0.0
- ReXeeD/Luminus-1.5B-Roleplay
- BeaverAI/Qwen2.5-QwQ-RP-Draft-v0.2-1.5B
- dphn/Dolphin3.0-Qwen2.5-1.5B
- Goekdeniz-Guelmez/Josiefied-Qwen2.5-1.5B-Instruct-abliterated-v1
- nbeerbower/EVA-abliterated-TIES-Qwen2.5-1.5B
- NathanRoll/writing-rlvr-qwen2.5-1.5b
base_model_relation: merge
library_name: gguf
pipeline_tag: text-generation
tags: [gguf, merge, qwen2.5, roleplay, creative-writing, fantasy, winery, llama.cpp, model-soup]
---
<p align="center"><img src="winery-logo.png" width="180"/></p>
🍷 Winery Qwen2.5 1.5B · RP Soup
An equal-parts model soup of eight Qwen2.5-1.5B roleplay, story and uncensored fine-tunes, blended with the Winery fusion compiler. It's built for fantasy roleplay and narration on phone-size hardware: 1.54B params, 1.65 GB at Q8_0.
Recipe
| Donor | Flavour |
|---|---|
| EVA-UNIT-01/EVA-Qwen2.5-1.5B-v0.0 | RP / story prose |
| EVA-UNIT-01/EVA-D-Qwen2.5-1.5B-v0.0 | RP prose, distilled |
| ReXeeD/Luminus-1.5B-Roleplay | character roleplay |
| BeaverAI/Qwen2.5-QwQ-RP-Draft-v0.2-1.5B | RP with reasoning |
| dphn/Dolphin3.0-Qwen2.5-1.5B | general, steerable |
| Goekdeniz-Guelmez/Josiefied-Qwen2.5-1.5B-Instruct-abliterated-v1 | stays in character (abliterated) |
| nbeerbower/EVA-abliterated-TIES-Qwen2.5-1.5B | EVA × abliterated |
| NathanRoll/writing-rlvr-qwen2.5-1.5b | RL-tuned creative writing |
name Winery RP Soup 1.5B
out q8_0
default linear 1,1,1,1,1,1,1,1
Not benchmarked: it's tuned for vibes, not MMLU. Try it in the Fantasy Tavern Space.
Use
Works in llama.cpp, Jan, LM Studio, SillyTavern (via llama-server/KoboldCpp) and the Winery app. Temperature 0.8–1.0 and min_p 0.05 work well.
llama-cli -m Winery-Qwen2.5-1.5B-RP-Soup-Q8_0.gguf -cnv
⚠️ Several donors are uncensored. It's meant for fiction between consenting adults, and you're responsible for how you use it.
Run WineryLabs/Winery-Qwen2.5-1.5B-RP-Soup-GGUF with guIDE
Download guIDE — the AI-native code editor with local LLM inference and 69 built-in tools.
Source: Hugging Face · Compare models