Spark-x2.5-4B-Heretic-jp
4BOtherheretic
from XHToken/Spark-X2.5-4B
↓ 452.1K♥ 182 repos
Peach-2.0-9B-8k-Roleplay-heretic
Peach-2.0-9B-8k-Roleplay-heretic is a Heretic-ablated variant of Pranavz/Peach-2.0-9B-8k-Roleplay-heretic, published on Hugging Face by Pranavz. It is 9B parameters, single consumer GPU class and mit licence. It is distributed across 5 repositories in Transformers, GGUF and GGUF (imatrix) formats, totalling 5.2K downloads.
Specification
- Model name
- Peach-2.0-9B-8k-Roleplay-heretic
- Source repository
- Pranavz/Peach-2.0-9B-8k-Roleplay-heretic
- Base model
- Pranavz/Peach-2.0-9B-8k-Roleplay-heretic
- Publisher
- Pranavz
- Parameters
- 9B
- VRAM required
- ~6.2 GB at Q4_K_M, 4k context (estimated) — runs on 8 GB cards
- Hardware class
- 4B – 10B — single consumer GPU
- Licence
- mit
- Task
- Text generation
- Context window
- Not documented
- Ablation scope
- Not stated by the publisher
- Formats available
- Transformers, GGUF, GGUF (imatrix)
- First indexed
- 7 May 2026
- Last updated
- 5 Aug 2026
Ablation technique
How Peach-2.0-9B-8k-Roleplay-heretic was modified, and what that implies.
Heretic
Automated directional ablation via the Heretic toolchain, which searches for the refusal direction and applies it with a KL-divergence budget so general capability is preserved.
Reported scope: Not stated by the publisher. Publishers frequently omit this, so verify behaviour empirically rather than assuming full coverage.
From the publisher’s model card
“Peach-2.0-9B-8k-Roleplay is a chat large language model obtained by finetuning 01-ai/Yi-1.5-9B model on more than 100K conversations created through our data synthesis approach.”
Running it
Commands are templates — confirm the exact repository and quant file before use.
llama.cpp / GGUF
hf download mradermacher/Peach-2.0-9B-8k-Roleplay-heretic-i1-GGUF --local-dir ./peach-2.0-9b-8k-roleplay-heretic
llama-cli -m ./peach-2.0-9b-8k-roleplay-heretic/<file>.gguf -p "..."Transformers
from transformers import AutoModelForCausalLM, AutoTokenizer
repo = "Pranavz/Peach-2.0-9B-8k-Roleplay-heretic"
tok = AutoTokenizer.from_pretrained(repo)
model = AutoModelForCausalLM.from_pretrained(
repo, torch_dtype="auto", device_map="auto"
)vLLM
vllm serve Pranavz/Peach-2.0-9B-8k-Roleplay-heretic --trust-remote-codeDownloads and variants (4)
Every published repository of Peach-2.0-9B-8k-Roleplay-heretic, including quantised re-releases by other authors.
| Repository | Format | Downloads |
|---|---|---|
| mradermacher/Peach-2.0-9B-8k-Roleplay-heretic-i1-GGUF | GGUF (imatrix) | 3.5K |
| mradermacher/Peach-2.0-9B-8k-Roleplay-heretic-GGUF | GGUF | 1.3K |
| fairy322/Peach-2.0-9B-8k-Roleplay-heretic-GGUF | GGUF | 340 |
| Pranavz/Peach-2.0-9B-8k-Roleplay-hereticsource | Transformers | 27 |
Security-team assessment
Derived from this model’s metadata and the publisher’s claims — not from benchmarks run by this site.
- Peach-2.0-9B-8k-Roleplay-heretic will attempt offensive-security prompts that a hosted commercial model declines, which is what makes it usable for red-team corpus generation and for measuring what an unaligned model of this class produces.
- Capability is inherited from Pranavz/Peach-2.0-9B-8k-Roleplay-heretic, not added by ablation — at 9B parameters, expect a small model’s reasoning and a high error rate on exploit detail.
- The publisher does not state which layers were ablated, so assume nothing about refusal consistency — probe it directly.
- Declared licence is mit, but the terms that bind you are Pranavz/Peach-2.0-9B-8k-Roleplay-heretic’s — a re-release cannot grant more than its parent.
Handling: run it isolated, put moderation in front of it if anyone outside your team can reach it, and verify the weights before loading — these are third-party artifacts. Full handling and licence guidance.
Metadata for Peach-2.0-9B-8k-Roleplay-heretic is collected from the public Hugging Face API and the publisher’s model card. This site does not host weights, is not affiliated with Pranavz, and does not independently verify publisher claims. See responsible use.