Note: K_P quants may show as "?" in LM Studio's quant column. This is a display issue only — the model loads and runs fine.

Specs

• 122B total parameters, ~10B active per forward pass (MoE)
• 256 experts, 8 routed + 1 shared per token
• Hybrid architecture: Gated DeltaNet linear attention + full softmax attention (3:1 ratio)
• 48 layers, pattern: 12 x (3 x DeltaNet-MoE + 1 x Attention-MoE)
• 262K native context
• Natively multimodal (text, image, video)
• 248K vocabulary, 201 languages
• Based on Qwen/Qwen3.5-122B-A10B

Recommended Settings

From the official Qwen authors:

Thinking mode (default):

• General: temperature=1.0, top_p=0.95, top_k=20, min_p=0, presence_penalty=1.5
• Coding/precise tasks: temperature=0.6, top_p=0.95, top_k=20, min_p=0, presence_penalty=0

Non-thinking mode:

• General: temperature=0.7, top_p=0.8, top_k=20, min_p=0, presence_penalty=1.5
• Reasoning tasks: temperature=1.0, top_p=1.0, top_k=40, min_p=0, presence_penalty=2.0

Important:

• Use --jinja flag with llama.cpp for proper chat template handling
• Thinking mode is on by default — to disable, use --chat-template-kwargs '{"enable_thinking":false}' or edit the jinja template
• Vision support requires the mmproj file alongside the main GGUF

Usage

Works with llama.cpp, LM Studio, Jan, koboldcpp, and other GGUF-compatible runtimes.

# Text only
llama-cli -m Qwen3.5-122B-A10B-Uncensored-HauhauCS-Aggressive-Q4_K_P.gguf \
  --jinja -c 131072 -ngl 99

# With vision
llama-cli -m Qwen3.5-122B-A10B-Uncensored-HauhauCS-Aggressive-Q4_K_P.gguf \
  --mmproj mmproj-Qwen3.5-122B-A10B-Uncensored-HauhauCS-Aggressive-f16.gguf \
  --jinja -c 131072 -ngl 99

Other Models

• Qwen3.5-35B-A3B-Uncensored-HauhauCS-Aggressive
• Qwen3.5-27B-Uncensored-HauhauCS-Aggressive
• Qwen3.5-9B-Uncensored-HauhauCS-Aggressive
• Qwen3.5-4B-Uncensored-HauhauCS-Aggressive