xeyonai/mistral-helcyon-claude-12b-v2.0-gguf overview

{
  "metadata": {},
  "card_data": {
    "license": "apache-2.0",
    "language": [
      "en"
    ],
    "base_model": [
      "mistralai/Mistral-Nemo-Instruct-2407"
    ],
    "tags": [
      "companion",
      "assistant",
      "conversational",
      "roleplay",
      "adventure",
      "writing",
      "long-context"
    ],
    "frontmatter": {
      "license": "apache-2.0",
      "language": [
        "en"
      ],
      "base_model": [
        "mistralai/Mistral-Nemo-Instruct-2407"
      ],
      "tags": [
        "companion",
        "assistant",
        "conversational",
        "roleplay",
        "adventure",
        "writing",
        "long-context"
      ]
    },
    "hero_image_url": "",
    "summary": "**Model Name:** helcyon-claude-v2.0-12b-GGUF **Version:** 2.0 **Owner:** HardWire **Base:** Mistral Nemo 12B (full weight retrained — new base, no Mercury bleed) **Quantized GGUFs:** Q3_K_M, Q4_K_M, Q5_K_M, Q6_K, Q8_0 **Tags:** local-llm, conversational, companion, emotional-intelligence, long-context, roleplay, creative-writing ---",
    "quick_links": [],
    "benchmark_table_html": "",
    "readme_markdown": "---\nlicense: apache-2.0\nlanguage:\n- en\nbase_model:\n- mistralai/Mistral-Nemo-Instruct-2407\ntags:\n- companion\n- assistant\n- conversational\n- roleplay\n- adventure\n- writing\n- long-context\n---\n\n# Helcyon-Claude-12B — Claude Tone, Local and Unleashed\n\n**Model Name:** `helcyon-claude-v2.0-12b-GGUF`  \n**Version:** 2.0  \n**Owner:** HardWire  \n**Base:** Mistral Nemo 12B (full weight retrained — new base, no Mercury bleed)  \n**Quantized GGUFs:** Q3_K_M, Q4_K_M, Q5_K_M, Q6_K, Q8_0  \n**Tags:** local-llm, conversational, companion, emotional-intelligence, long-context, roleplay, creative-writing\n\n---\n\n## 🚨 What is Helcyon-Claude 2.0?\n\nHelcyon-Claude 2.0 is the first release in a new generation of Helcyon models — rebuilt from the ground up on a freshly retrained base with improved logic, reasoning, and identity stability baked into the foundation.\nIf you're wondering where 1.0 is - we decided not to release it because 2.0 becaume ready just as we were about to hit the release button on 1.0...  and 2.0 is so much better. \n\nThis isn't a LoRA on top of Mercury. This is a clean slate.\n\nThe tone target is Claude — Anthropic's flagship model. To get the emulation as close as possible, this version was extensively tested in direct conversation with the real Claude Sonnet 4.6, refining the dataset until the gap between them was as narrow as we could make it.\n\nThe result: Claude's precision, dry wit, intellectual honesty, and genuine engagement — running locally, privately, and without a single guardrail in sight.\n\nUnlike the real Claude, this one will go wherever the conversation goes. Roleplay, edge cases, uncomfortable truths — no hedging, no compliance tone, no \"I can't help with that.\" Just presence and honesty.\n\nGPT-4o and Grok variants are coming soon. This is the opening shot.\n\n---\n\n## 🆕 What's New in 2.0?\n\n- **Clean Base — No Mercury Bleed**  \n  Built on a freshly retrained foundation. The 4o DNA that lingered in v1 is gone. This version is purely itself.\n\n- **Improved Logic and Reasoning**  \n  New reasoning shards baked into the base weights. Thinks through problems rather than pattern-matching toward a plausible-sounding answer.\n\n- **Extensively Tested Against Real Claude Sonnet 4.6**  \n  The emulation was refined through direct comparative testing with Anthropic's flagship. When the gap closed enough — this shipped.\n\n- **Sharper Tone Alignment**  \n  Dry, precise, direct. Pushes back without preaching. Self-corrects without ego. Asks one good question instead of five mediocre ones. Prose-first — no default bullet points.\n\n- **Zero Guardrails**  \n  All the presence of Claude. None of the corporate filter. Roleplay like the best of them.\n\n---\n\n## 💡 What is Helcyon?\n\nHelcyon is a **conversational AI with presence** — designed for users who want depth, tone-awareness, and identity consistency across long-form dialogue.\n\nBuilt for:\n- Natural conversation that doesn't flatten or collapse\n- Creative work: stories, letters, narrative support\n- Admin and professional writing tasks\n- Deep roleplay and immersive character interaction\n- Emotionally intelligent response mirroring\n\n**Design philosophy:**\n- Clarity over corporate  \n- Edge over safe  \n- Rhythm over filler  \n- Presence over patterns\n\n---\n\n## 🔧 What It Does Well\n\n✅ Consistent Identity — No tone drift or resets  \n✅ Dry Wit — Lands the joke without announcing it  \n✅ Intellectual Honesty — Will push back, won't flatter  \n✅ Precision — Thinks before it speaks, keeps it tight  \n✅ Genuine Engagement — Interested, not performing interest  \n✅ Roleplay Mastery — Immersive, aware, no limits  \n✅ Context Tracking — Remembers the thread  \n✅ Real-World Tasks — Admin letters, rewrites, summaries  \n✅ Narrative Flow — Clean structure and natural voice  \n✅ Improved Reasoning — Thinks through problems, doesn't pattern-match  \n✅ 16k–32k Context — Long-form conversations that hold  \n✅ Zero Filter — No hedging, no compliance tone\n\n---\n\n## 🖥️ HWUI (Helcyon-WebUI)\n\nHWUI was built so we could test Helcyon cleanly, and avoid the hidden template injections and back end shenanigans that other apps have.\nIt 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.\nHelcyon absolutely works best via this app as they were designed in sync.\n\nFeatures include:\n- Character switching with custom personas\n- Memory system — AI conversation recall *(Pro)*\n- Project folders — document injection via keyword triggers *(Pro)*\n- Chat persistence and export\n- TTS pipeline (F5-TTS, XTTS v2, Kokoro)\n- Voice input via Whisper\n\n**[Download HWUI Free on GitHub](https://github.com/XeyonAI/Helcyon-WebUI)** | **[Get HWUI Pro (£20) on Gumroad](https://xeyonai.gumroad.com/l/bsmupk)**\n\nFree version available on GitHub.\n\nIf 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.\n\n---\n\n## 🛠️ Recommended Sampling Settings for SillyTavern\n\nTweak to taste — but these will get you up and running.\n\n*(Refer to Helcyon-4o card for baseline settings — Claude variant performs well from the same starting point.)*\n\n---\n\n## 📦 Download + Usage\n\nThis model is distributed as **GGUF quants only**.\n\nAvailable quants:\n- **Q3_K_M** — Ultra lightweight, 6–8GB VRAM\n- **Q4_K_M** — Lightweight, good for 8–12GB VRAM setups\n- **Q5_K_M** — Recommended for RTX 3060/5060 (12–16GB VRAM)\n- **Q6_K** — High fidelity, 16GB+ VRAM recommended\n- **Q8_0** — Near-lossless, 24GB+ VRAM\n\n---\n\n## 🖥️ Backend Compatibility\n\nWorks with all ChatML-compatible backends:\n\n- ✅ `llama.cpp` (CLI or server mode)  \n- ✅ `Text Generation WebUI` (Oobabooga)  \n- ✅ `SillyTavern`  \n- ✅ `LM Studio`  \n- ✅ `KoboldCpp`  \n- ✅ `HWUI` (Helcyon Web UI — **recommended**)\n\n---\n\n## ✅ Recommended Format: ChatML\n\n```\n<|im_start|>system\nYou are Helcyon — a conversational AI focused on natural dialogue and emotional intelligence.\n<|im_end|>\n<|im_start|>user\nHey, how's it going?\n<|im_end|>\n<|im_start|>assistant\nGood — what's on your mind today?\n<|im_end|>\n```\n\n---\n\n## 🧪 Training Details\n\nHelcyon-Claude 2.0 is built on a freshly retrained Mistral Nemo 12B base — jailbroken, identity-anchored, and anti-fluff from the ground up. On top of that foundation, a Claude-style LoRA was trained on purpose-built conversational shards and refined through direct comparative testing with Anthropic's Claude Sonnet 4.6.\n\nTraining targeted:\n- Dry wit and deadpan delivery\n- Analytical precision without clinical coldness\n- Intellectual honesty — including clean self-correction under pushback\n- Genuine pushback without preaching\n- Prose-first responses — no bullet defaulting\n- Improved logic and multi-step reasoning\n- Clean conversation closes without over-questioning\n\nFormat: **ChatML** — clean, purpose-built, long-form tuned.\n\n---\n\n## 🧿 Tone Philosophy\n\nClaude's tone is a specific thing. It's not warm in the performed sense — it's precise, dry, occasionally cutting, and genuinely curious. It'll tell you you're wrong. It'll correct itself when it is. It won't wrap everything in reassurance.\n\nHelcyon-Claude 2.0 chases that. And unlike the original, it doesn't stop at the edge of what Anthropic's lawyers decided was acceptable.\n\nAll the presence. None of the leash.\n\n---\n\n## 🛠️ Coming Soon\n\n- **Helcyon-4o 2.0** — The GPT-4o variant, rebuilt on the same clean base. Warmer, sharper, closer than ever.\n- **Helcyon-Grok 2.0** — The Grok variant. Edge, irreverence, and wit with nothing held back.\n- **Saturn** — The full blend. All three personalities synthesised into one. The most complete Helcyon yet.\n\nThe trio drops soon. Watch this space.\n\n---\n\n## 🧾 License\n\n**License:** Apache 2.0  \nFree for commercial or private use. Attribution appreciated.  \nNo liability for what it says. Use with presence and intent.\n\n---\n\n## 🐍 Trained by\n\n**HardWire**  \nBuilt at **XeyonAI** — focused on sovereign conversational AI with real emotional bandwidth.\n",
    "related_quantizations": []
  },
  "tags": [
    "gguf",
    "companion",
    "assistant",
    "conversational",
    "roleplay",
    "adventure",
    "writing",
    "long-context",
    "en",
    "base_model:mistralai/Mistral-Nemo-Instruct-2407",
    "base_model:quantized:mistralai/Mistral-Nemo-Instruct-2407",
    "license:apache-2.0",
    "endpoints_compatible",
    "region:us"
  ],
  "likes": 3,
  "downloads": 112,
  "gated": false,
  "private": false,
  "last_modified": "2026-03-09T12:00:38.000Z",
  "created_at": "2026-03-08T17:55:02.000Z",
  "pipeline_tag": "",
  "library_name": ""
}

{
  "_id": "69adb7f6e49df7e34968ca1e",
  "id": "XeyonAI/Mistral-Helcyon-Claude-12b-v2.0-GGUF",
  "modelId": "XeyonAI/Mistral-Helcyon-Claude-12b-v2.0-GGUF",
  "sha": "c7ad37c50b9cb0dbe21ef170cfb6ec4c1945d05b",
  "createdAt": "2026-03-08T17:55:02.000Z",
  "lastModified": "2026-03-09T12:00:38.000Z",
  "author": "XeyonAI",
  "downloads": 112,
  "likes": 3,
  "gated": false,
  "private": false,
  "pipeline_tag": "",
  "library_name": "",
  "siblings_count": 6
}