---

base_model:

• EryriLabs/Ornith-1.5-35B-A3B-BigBang-MTP-GGUF
• Kwaipilot/KAT-Coder-V2.5-Dev
• Accio-Lab/occamy-1.0
• AllSpark-Research/CompoWorld

base_model_relation: merge

library_name: transformers

pipeline_tag: text-generation

tags:

• qwen3.5
• agent
• coding
• code
• kat-coder
• ornith
• occamy
• compoworld
• reasoning
• tool-use
• safetensors
• gguf
• apex
• imatrix
• llama.cpp

---

Qwen3.6-Qworus-35B-A3B-APEX-GGUF

What is this?

This is a DELLA merge of the following finetuned & post-trained models:

• EryriLabs/Ornith-1.5-35B-A3B-BigBang-MTP-GGUF: The base model, for terminal use, coding, software architecture, planning. Also provides MTP support.
• Kwaipilot/KAT-Coder-V2.5-Dev: Contributes efficient coding with fewer tokens over more languages.
• Accio-Lab/occamy-1.0: Contributes general agent abilities, also a semi-official finetune.
• AllSpark-Research/CompoWorld: Also contributes general agent abilities, different strengths/weaknesses from occamy.

The result is an intelligent and efficient hybrid model for coding, terminal agent work, planning, designing, and conversation.

Although most of this model's weights were merged with with a method that drops weights, more sensitive layers such as the router gates and shared experts, were protected by using a density of 1.0 and epsilon of 0.0 to disable sparsification.

Versions:

V1.2: New, further revised DELLA recipe which is both smarter and more stable than V1.1, but with somewhat more verbose reasoning. Besides changing parameters, I also added lm_head and embed_tokens to the layers protected from sparsification.

V1.1: Revised DARE-TIES recipe which brings in the CompoWorld 35B-A3B finetune and gives more ratio to KAT-Coder and less towards Ornith; this version is significantly better at coding and has a wider spread of abilities.

V1: Initial DARE-TIES version which is good at terminal use, but not as proficient in coding tasks.

---

Quantization

Weights are provided quantized with APEX, which preserves sensitive MoE tensors. I-Quality is the best size for high quality coding and reasoning, but I may add other sizes later.

The importance matrix was calibrated on about 50% random English text snippets from multiple domains, and 50% tokens generated during real long-context agent sessions with KAT-Coder and Ornith-1.5 (control tokens such as <|im_start|> are included).