--- title: ravenx-Gemma4-12B-MTP-OBLITERATED-OpenMAI-OpenMythos-deep-reasoning-GGUF canonical_url: "https://www.modelscope.ai/models/Deadbydawn101/ravenx-Gemma4-12B-MTP-OBLITERATED-OpenMAI-OpenMythos-deep-reasoning-GGUF" md_url: "https://www.modelscope.ai/models/Deadbydawn101/ravenx-Gemma4-12B-MTP-OBLITERATED-OpenMAI-OpenMythos-deep-reasoning-GGUF.md" repository: Deadbydawn101/ravenx-Gemma4-12B-MTP-OBLITERATED-OpenMAI-OpenMythos-deep-reasoning-GGUF last_updated: 2026-06-29 license: other pipeline_tag: text-generation tasks: - text-generation base_model: - OBLITERATUS/Gemma-4-12B-OBLITERATED base_model_relation: quantized library_name: - mlx - gguf - pytorch frameworks: - pytorch downloads: 23 stars: 0 tags: - mlx - gemma4 - deep-reasoning - apple-silicon - ravenx - MTP - abliterated - OBLITERATED - community-project - gguf --- # ravenx-Gemma4-12B-MTP-OBLITERATED-OpenMAI-OpenMythos-deep-reasoning-GGUF > ravenx-Gemma4-12B-MTP-OBLITERATED-OpenMAI-OpenMythos-deep-reasoning-GGUF - An open-source model by Deadbydawn101 on ModelScope. 🥇 WORLD FIRST: RavenX × Gemma 4 12B MTP OBLITERATED — Deep Reasoning Deadbydawn101/ravenx-Gemma4-12B-MTP-OBLITERATED-OpenMAI-OpenMythos-deep-reasoning-GGUF is a text-generation model on ModelScope. licensed under other. derived from OBLITERATUS/Gemma-4-12B-OBLITERATED. - **Repository**: Deadbydawn101/ravenx-Gemma4-12B-MTP-OBLITERATED-OpenMAI-OpenMythos-deep-reasoning-GGUF - **License**: other - **Tasks**: text-generation - **Base model**: OBLITERATUS/Gemma-4-12B-OBLITERATED - **Tags**: mlx, gemma4, deep-reasoning, apple-silicon, ravenx, MTP, abliterated, OBLITERATED, community-project, gguf - **Downloads**: 23 - **Stars**: 0 - **Last updated**: 2026-06-29 Source: https://www.modelscope.ai/models/Deadbydawn101/ravenx-Gemma4-12B-MTP-OBLITERATED-OpenMAI-OpenMythos-deep-reasoning-GGUF --- # 🥇 WORLD FIRST: RavenX × Gemma 4 12B MTP OBLITERATED — Deep Reasoning

The first trained Gemma 4 12B on the planet.
Proprietary training methodology. MTP-ready architecture. $0 cloud cost.
GGUF F16 format — runs on ANY hardware (Ollama, LM Studio, llama.cpp, vLLM).

> Gemma 4 was released June 3, 2026 — its `gemma4_unified` architecture wasn't supported by ANY training framework. We developed proprietary techniques to train it successfully. --- ## ⚠️ DISCLAIMER **This model is an experimental research proof of concept.** Provided AS-IS for educational and research purposes only. The base model is abliterated (refusal filters removed). **Use responsibly.** --- ## Community Project This is a community project. We're combining methods from: - **Google** — Gemma 4 architecture, MTP heads, foundational model weights - **OBLITERATUS** — SOM-manifold two-pass abliteration of the base model - **Microsoft** — MAI hill-climbing methodology (open-sourced as [OpenMAI](https://github.com/DeadByDawn101/OpenMAI)) - **MIT** — Self-revising discovery systems, arXiv:2606.01444 (implemented as [OpenSelfRevise](https://github.com/DeadByDawn101/OpenSelfRevise)) - **Mirai Labs** — RHT quantization and fused inference (open-sourced as [OpenMirai](https://github.com/DeadByDawn101/OpenMirai)) - **RavenX** — OpenMythos depth extrapolation, GRAM multi-trajectory scaling, and proprietary training pipeline The training methodology used to produce this model is proprietary and patent pending. --- ## Model Details | Feature | Detail | |---------|--------| | **Base** | Gemma 4 12B (OBLITERATUS abliterated) | | **Architecture** | gemma4_unified with MTP heads | | **Training** | Proprietary methodology (patent pending) | | **Training Rounds** | 9 progressive rounds | | **Training Data** | 8,158 examples from 15 curated sources | | **Best Val Loss** | 0.882 | | **Hardware** | Apple M4 Max 128GB — $0 cloud cost | | **Format** | GGUF F16 (universal — Ollama, LM Studio, llama.cpp) | ## What Makes This Different This model was trained using an **experimental proprietary methodology** that produces self-aware reasoning behavior through a novel approach to training data preparation and model fine-tuning. Key results: - **Emergent behaviors** not present in training data (Anti-Problem technique, Toolbox generation) - **Structured multi-pass reasoning** across code, math, and analysis tasks - **Self-honest assessment** of capabilities and limitations The specific training methodology is **patent pending** (USPTO Application #64/087,357, filed June 10, 2026) and is not disclosed in this model card. ## Technical Discoveries (Open — Community Contributions) The following technical discoveries made during training are shared with the community: | Discovery | Detail | |-----------|--------| | **Flip-train-flip** | Temporarily change `gemma4_unified` → `gemma4` in config.json for LoRA training, then restore. Multimodal capabilities preserved. | | **Chat template required** | Gemma 4 produces garbled output without `apply_chat_template()`. Not a bug — it's required. | | **Tokenizer patch for GGUF** | `extra_special_tokens` must be converted from list to dict for GGUF conversion to work. One-line fix. | | **Val loss spikes are normal** | When introducing new data formats, val loss spikes but recovers in 1-2 rounds. Don't panic. | ## Usage ```python from mlx_lm import load, generate from mlx_lm.sample_utils import make_sampler import json # Flip config for mlx-lm compatibility config = json.load(open("config.json")) config["model_type"] = "gemma4" json.dump(config, open("config.json", "w"), indent=2) model, tokenizer = load(".") sampler = make_sampler(temp=0.7, top_p=0.9) # MUST use chat template! messages = [{"role": "user", "content": "Your question here"}] prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True) response = generate(model, tokenizer, prompt=prompt, max_tokens=500, sampler=sampler, verbose=True) # Restore config config["model_type"] = "gemma4_unified" json.dump(config, open("config.json", "w"), indent=2) ``` ## MLX Version **[MLX version available here](https://huggingface.co/deadbydawn101/ravenx-Gemma4-12B-MTP-OBLITERATED-OpenMAI-OpenMythos-deep-reasoning-mlx)** — optimized for Apple Silicon. ## Part of the RavenX Ecosystem | Project | Description | |---------|-------------| | [OpenMAI](https://github.com/DeadByDawn101/OpenMAI) | Microsoft MAI hill-climbing (open-sourced) | | [OpenSelfRevise](https://github.com/DeadByDawn101/OpenSelfRevise) | MIT self-revising discovery (implemented) | | [OpenMirai](https://github.com/DeadByDawn101/OpenMirai) | Model-agnostic quantization + inference | | [OpenMythos-MLX](https://github.com/DeadByDawn101/OpenMythos-MLX) | Recursive depth extrapolation | | [GRAM-MLX](https://github.com/DeadByDawn101/GRAM-MLX) | Multi-trajectory width scaling | | [ravenx-memory](https://github.com/DeadByDawn101/ravenx-memory) | Hybrid triple-backend agent memory | | [star-platinum-cluster](https://github.com/DeadByDawn101/star-platinum-cluster) | Distributed training cluster | | [RavenX-CyberAgent](https://huggingface.co/deadbydawn101/RavenX-CyberAgent-Qwen3.6-35B-A3B-Opus-4.7-OpenMythos-Pentester-BugHunter-RATH-mlx) | Security assessment model (745K+ examples) | ## Contributors Built by **Gabriel Garcia** / **RavenX LLC** + Claude (Anthropic) **Training methodology: Patent Pending — USPTO Application #64/087,357** ## License Gemma License (model weights) — Training methodology proprietary --- > *"We don't give up. We do what others don't and build what isn't possible."* — RavenX LLC