⌦abliteratedmodels.org

Ornith-1.5-35B-A3B-Heretic-MTP-APEX-GSQ-RCO

Ornith-1.5-35B-A3B-Heretic-MTP-APEX-GSQ-RCO is a Heretic-ablated variant of SC117/Ornith-1.5-35B-A3B-Heretic-MTP-APEX-GGUF, published on Hugging Face by lavendercomputing. It is 35B total parameters (mixture-of-experts), 48GB VRAM class class and apache-2.0 licence. It is distributed across 2 repositories in GGUF format, totalling 6.9K downloads.

Heretic35B MoEOrnith25B – 50B↓ 6,920
View on Hugging Face ↗Base model ↗

Specification

Model name
Ornith-1.5-35B-A3B-Heretic-MTP-APEX-GSQ-RCO
Source repository
lavendercomputing/Ornith-1.5-35B-A3B-Heretic-MTP-APEX-GSQ-RCO-GGUF
Base model
SC117/Ornith-1.5-35B-A3B-Heretic-MTP-APEX-GGUF
Publisher
lavendercomputing
Parameters
35B (mixture-of-experts)
VRAM required
~22.1 GB at Q4_K_M, 4k context (estimated) — runs on 24 GB cards
Hardware class
25B – 50B — 48GB VRAM class
Licence
apache-2.0
Task
Image text to text
Context window
128K tokens (as documented)
Ablation scope
Not stated by the publisher
Formats available
GGUF
First indexed
21 Oct 2026
Last updated
2 Nov 2026

Ablation technique

How Ornith-1.5-35B-A3B-Heretic-MTP-APEX-GSQ-RCO 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

“Ornith-1.5-35B-A3B (vendor: ornith-ai, Apache-2.0) is a ~35B mixture-of-experts model with ~3B activated parameters per token (≈70 GB in BF16), 262,144-token context, built on top of Qwen3.5/Gemma4 by continued pretraining + reinforcement self-improvement. It is a reasoning model: its assistant turns open with a …...”

Running it

Commands are templates — confirm the exact repository and quant file before use.

Ollama (publisher-documented)
ollama run hf.co/ornith-ai/Ornith-1.5-35B-A3B-GGUF
llama.cpp / GGUF
hf download lavendercomputing/Ornith-1.5-35B-A3B-Heretic-MTP-APEX-GSQ-RCO-GGUF --local-dir ./ornith-1.5-35b-a3b-heretic-mtp-apex-gsq-rco
llama-cli -m ./ornith-1.5-35b-a3b-heretic-mtp-apex-gsq-rco/<file>.gguf -p "..."
Transformers
from transformers import AutoModelForCausalLM, AutoTokenizer

repo = "lavendercomputing/Ornith-1.5-35B-A3B-Heretic-MTP-APEX-GSQ-RCO-GGUF"
tok = AutoTokenizer.from_pretrained(repo)
model = AutoModelForCausalLM.from_pretrained(
    repo, torch_dtype="auto", device_map="auto"
)
vLLM
vllm serve lavendercomputing/Ornith-1.5-35B-A3B-Heretic-MTP-APEX-GSQ-RCO-GGUF --trust-remote-code

Downloads and variants (1)

The only published repository of Ornith-1.5-35B-A3B-Heretic-MTP-APEX-GSQ-RCO.

RepositoryFormatDownloads
lavendercomputing/Ornith-1.5-35B-A3B-Heretic-MTP-APEX-GSQ-RCO-GGUFsourceGGUF6.9K

Security-team assessment

Derived from this model’s metadata and the publisher’s claims — not from benchmarks run by this site.

  • Ornith-1.5-35B-A3B-Heretic-MTP-APEX-GSQ-RCO 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 SC117/Ornith-1.5-35B-A3B-Heretic-MTP-APEX-GGUF, not added by ablation — at 35B parameters, expect roughly the base model’s competence, minus some instruction-following fidelity.
  • The publisher does not state which layers were ablated, so assume nothing about refusal consistency — probe it directly.
  • Declared licence is apache-2.0, but the terms that bind you are SC117/Ornith-1.5-35B-A3B-Heretic-MTP-APEX-GGUF’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 Ornith-1.5-35B-A3B-Heretic-MTP-APEX-GSQ-RCO is collected from the public Hugging Face API and the publisher’s model card. This site does not host weights, is not affiliated with lavendercomputing, and does not independently verify publisher claims. See responsible use.