CoolFace
Modelpublic

jakeroxs/KAT-Coder-V2.5-Dev-35B-A3B-MTP-ABLITERATED

sourceHugging Faceapache-2.0updated 24d agoView on Hugging Face
0likes269downloads
Model Card

KAT-Coder V2.5 Dev 35B-A3B MTP Abliterated

A merged Hugging Face-format checkpoint combining the Philadelphia Class abliteration of KAT-Coder V2.5 Dev 35B-A3B with the compatible Multi-Token Prediction (MTP) layer from KAT-Coder V2.5 Dev.

This checkpoint is the source model used to produce the GGUF versions available here:

[jakeroxs/KAT-Coder-V2.5-Dev-35B-A3B-MTP-ABLITERATED-GGUF](https://huggingface.co/jakeroxs/KAT-Coder-V2.5-Dev-35B-A3B-MTP-ABLITERATED-GGUF)

Model lineage

Original model

KwaiPilot/KAT-Coder-V2.5-Dev

Abliterated base

KridgeDookie/KAT-Coder-V2.5-Dev-35B-A3B-ABLITERATED-UNCENSORED-PHILADELPHIA-CLASS

MTP head

Myric/KAT-Coder-V2.5-Dev-MTP-head

The Philadelphia Class checkpoint is used as the target model. The compatible MTP tensors from Myric's KAT-Coder V2.5 Dev MTP-head checkpoint were transplanted into it.

Conceptually:

text
KwaiPilot/KAT-Coder-V2.5-Dev
             ↓
KridgeDookie/...PHILADELPHIA-CLASS
             ↓
       + Myric MTP head
             ↓
jakeroxs/...MTP-ABLITERATED
             ↓
         GGUF conversion

MTP modification

The Philadelphia Class checkpoint originally has:

text
mtp_num_hidden_layers = 0

This merged checkpoint changes that to:

text
mtp_num_hidden_layers = 1

and adds the 19 MTP tensors from the KAT-Coder V2.5 Dev MTP-head checkpoint.

The resulting weight layout is:

FileTensor count
model.safetensors693
model-00014-of-mtp.safetensors19
Total712

The accompanying model.safetensors.index.json maps the original model weights and transplanted MTP tensors to their respective files.

Architecture

PropertyValue
ArchitectureQwen3.5 MoE / qwen35moe
Parameters~35.51B
Active parameters~3B
Experts256
Experts used per token8
Native context262,144 tokens
MTP prediction layers1

Intended use

This checkpoint is primarily provided for:

  • —further conversion or quantization
  • —llama.cpp GGUF generation
  • —experimentation with the transplanted MTP layer
  • —local coding and agentic coding workflows
  • —research into speculative decoding with KAT-Coder V2.5 Dev

For direct llama.cpp use, the GGUF repository is generally more convenient:

[KAT-Coder-V2.5-Dev-35B-A3B-MTP-ABLITERATED-GGUF](https://huggingface.co/jakeroxs/KAT-Coder-V2.5-Dev-35B-A3B-MTP-ABLITERATED-GGUF)

About the abliteration

This model inherits the modifications made by the Philadelphia Class checkpoint.

Abliteration attempts to reduce refusal behavior by modifying model representations associated with refusal behavior. It does not retrain the underlying model and does not guarantee any particular behavior.

The underlying coding capabilities originate from KAT-Coder V2.5 Dev.

Credits

  • —KwaiPilot — KAT-Coder V2.5 Dev
  • —KridgeDookie — Abliterated / Uncensored Philadelphia Class variant
  • —Myric — KAT-Coder V2.5 Dev MTP head

This repository packages the merged Hugging Face-format checkpoint used as the source for the associated GGUF release.