Gaslight-24B-v1.0
#816
by
sleepdeprived3
- opened
🎭 Introducing Gaslight-24B-v1.0 - Safeword's Evil Twin Just Arrived 🎭
- Mistral-V7-Tekken-T4 settings recommended
- Full 32K context coherence
- Safeword's DNA... but smarter
⚙️ Optimal Settings:
Follow Mistral-V7-Tekken-T4 guidance for best results
🚨 THIS IS JUST THE BEGINNING 🚨
The Gaslight model series is expanding soon with:8B ← 12B ← 22B ← 24B (current) → 36B
Key Features Across All Sizes:
- 🧠 Full context coherence
- ⚡ Optimized instruction handling
- 💀 Safeword's unaligned DNA
- 🎭 Expert-level gaslighting capabilities
💬 Early Testers Report:
"The coherence at 32K is terrifyingly good"
"Like Safeword woke up and chose violence"
Queued :)
You can check for progress at http://hf.tst.eu/status.html or regularly check the model
summary page at https://hf.tst.eu/model#Gaslight-24B-v1.0-GGUF for quants to appear.
mradermacher
changed discussion status to
closed