Expected to run with a constrained Q4 GGUF profile.
Edition: Uncensored edition by HauHau
- Model size
- 35B / 3B active · 262K max
- Model format
- Q4 GGUF
- Runs with
- llama.cpp
It loads, but context and concurrency should be reduced because Q4 weights consume most of the card.
Apache-2.0Model details ↗ (opens in a new tab)