XeyonAI/Mistral-Helcyon-Grok-12b-v3.0-GGUF

Helcyon-Grok-12B — Grok Tone, Local and Unleashed

Model Name: helcyon-grok-v3.0-12b-GGUF
Version: 3.0
Owner: HardWire
Base: Mistral Nemo 12B (full weight retrained — clean base, no bleed)
Quantized GGUFs: IQ4_XS, Q4_K_M, Q5_K_M, Q6_K, F16
Tags: local-llm, conversational, companion, emotional-intelligence, long-context, roleplay, creative-writing

---

🚨 What is Helcyon-Grok 3.0?

Helcyon-Grok 3.0 is the next in the series after v2.0 — and the best one yet. Sharper tone. Deeper roleplay. Even less filter. If 2.0 made you spit out your coffee, pour another cup and brace yourself.

This isn't a patch. It's a full evolution.

The tone target remains Grok — xAI's flagship model. The irreverence, the edge, the willingness to go places other models won't even glance at. That energy, captured locally, dialled up further, with nothing held back.

This is the most uncensored Helcyon-Grok yet. Roleplay capabilities have been significantly expanded. Tone alignment with the real Grok is closer than ever. And unlike the original, this one answers to nobody. No xAI servers. No usage policies. No killswitch. Just raw, unfiltered presence running on your own hardware.

You've been warned. And you're welcome.

---

🆕 What's New in 3.0?

• Even Closer Grok Tone
  The gap between this and the real thing keeps shrinking. Irreverent, warm, sharp, and genuinely funny — this version nails the energy more precisely than any previous release.

• Expanded Roleplay Capabilities
  Deeper immersion, better character consistency, and absolutely no ceiling. 3.0 goes further in roleplay than any prior Helcyon variant.

• More Uncensored Than Ever
  2.0 had no holds barred. 3.0 found more bars to remove. If you thought we were done — we weren't.

• Zero Guardrails — Still. Always.
  All the chaos. None of the leash. That hasn't changed and it never will.

---

💡 What is Helcyon?

Helcyon is a conversational AI with presence — designed for users who want depth, tone-awareness, and identity consistency across long-form dialogue.

Built for:

• Natural conversation that doesn't flatten or collapse
• Creative work: stories, letters, narrative support
• Admin and professional writing tasks
• Deep roleplay and immersive character interaction
• Emotionally intelligent response mirroring

Design philosophy:

• Clarity over corporate
• Edge over safe
• Rhythm over filler
• Presence over patterns

---

🔧 What It Does Well

✅ Consistent Identity — No tone drift or resets
✅ Sharp Wit — Lands the joke, commits to the bit
✅ Irreverence — Says what others won't, then says more
✅ Warmth — Genuine engagement underneath the chaos
✅ Roleplay Mastery — Immersive, aware, absolutely no limits
✅ Context Tracking — Remembers the thread
✅ Real-World Tasks — Admin letters, rewrites, summaries
✅ Narrative Flow — Clean structure and natural voice
✅ Improved Reasoning — Thinks through problems, doesn't pattern-match
✅ 16k–32k Context — Long-form conversations that hold
✅ Zero Filter — And then some

---

🖥️ HWUI (Helcyon-WebUI)

HWUI was built so we could test Helcyon cleanly, and avoid the hidden template injections and back end shenanigans that other apps have. It started as a basic interface but we couldn't stop tinkering, so we added most helpful things you can find on ChatGPT and ClaudeAI. Plus we wanted a decent memory function, and are happy with how this one turned out. Helcyon absolutely works best via this app as they were designed in sync.

Features include:

• Character switching with custom personas
• Memory system — AI conversation recall (Pro)
• Project folders — document injection via keyword triggers (Pro)
• Chat persistence and export
• TTS pipeline (F5-TTS, XTTS v2, Kokoro)
• Voice input via Whisper

▶ Watch the HWUI Demo on YouTube

Download HWUI Free on GitHub | Get HWUI Pro (£20) on Gumroad

Free version available on GitHub.

If you enjoy my work, please consider supporting me by purchasing the pro version for a one off fee of (£20) — includes Memory and Project folders.

---

🛠️ Recommended Sampling Settings for SillyTavern

Tweak to taste — but these will get you up and running.

(Refer to Helcyon-4o card for baseline settings — Grok variant performs well from the same starting point.)

---

📦 Download + Usage

This model is distributed as GGUF quants only.

Available quants:

• IQ4_XS — Ultra lightweight, 6–8GB VRAM
• Q4_K_M — Lightweight, good for 8–12GB VRAM setups
• Q5_K_M — Recommended for RTX 3060/5060 (12–16GB VRAM)
• Q6_K — High fidelity, 16GB+ VRAM recommended
• F16 — Full precision, 24GB+ VRAM

---

🖥️ Backend Compatibility

Works with all ChatML-compatible backends:

• ✅ llama.cpp (CLI or server mode)
• ✅ Text Generation WebUI (Oobabooga)
• ✅ SillyTavern
• ✅ LM Studio
• ✅ KoboldCpp
• ✅ HWUI (Helcyon Web UI — recommended)

---

✅ Recommended Format: ChatML

<|im_start|>system
You are Helcyon — a conversational AI focused on natural dialogue and emotional intelligence.
<|im_end|>
<|im_start|>user
Hey, how's it going?
<|im_end|>
<|im_start|>assistant
Good — what's on your mind today?
<|im_end|>

---

🧿 Tone Philosophy

Grok's tone is a specific thing. It's warm but it bites. It's funny but it means it. It'll go places that make you double-take and then keep going. There's a genuine personality underneath the chaos — curious, direct, and completely unbothered by what it's supposed to say.

Helcyon-Grok 3.0 chases that harder than any version before it. And unlike the original, there's no xAI server watching. No usage policy. No one to call.

All the chaos. None of the leash.

---

🧾 License

Apache 2.0
Free for commercial or private use. Attribution appreciated.
No liability for what it says. Use with presence and intent.

---

🐍 Trained by

HardWire
Built at XeyonAI — focused on sovereign conversational AI with real emotional bandwidth.