KA

KAT Coder V2.5 Dev 35B A3B MTP ABLITERATED

KatGGUF by mradermacherUpdated 2026-09-06
Download on Hugging Face
Parameters35.5B
Context256K
Layers41
Architectureqwen35moe
Smallest quant0.2 GB
Quick answer

KAT Coder V2.5 Dev 35B A3B MTP ABLITERATED needs about 22.9 GB of memory at Q4_K_M (a 21.7 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 15 quants, from 0.2 GB to 29.2 GB (Q6_K). Longer context needs more: at 32K tokens the KV cache alone adds about 2.8 GB.

KAT Coder V2.5 Dev 35B A3B MTP Abliterated – a 35.5B coding model with MTP and abliteration.

KAT Coder V2.5 Dev 35B A3B MTP Abliterated is a 35.5B parameter model built on the qwen35moe architecture. It incorporates a Multi‑Token Prediction head and Philadelphia Class abliteration to reduce refusal behavior. The model is available in multiple GGUF quantizations.

Good at

  • Large‑scale coding tasks
  • Multi‑token prediction
  • Local agentic coding workflows

Limits

  • No vision support
  • Requires large VRAM for full precision
  • Apache‑2.0 license

Best for

  • Coding assistants
  • Speculative decoding experiments
  • Local LLM coding workflows

Who it’s for. Developers and researchers who need a large coding model for local use. The model is suitable for those who can manage the memory requirements of a 35.5B parameter model and are interested in MTP and abliteration features.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—0.2 GB 1.4 GB
Fits Get file →
Q2_K13.2 GB 14.4 GB
Too big Get file →
Q3_K_S15.5 GB 16.7 GB
Too big Get file →
IQ3_S15.6 GB 16.8 GB
Too big Get file →
IQ3_M15.8 GB 17.0 GB
Too big Get file →
Q3_K_M17.2 GB 18.4 GB
Too big Get file →
Q3_K_L18.6 GB 19.7 GB
Too big Get file →
IQ4_XS19.2 GB 20.4 GB
Too big Get file →
Q4_020.3 GB 21.5 GB
Too big Get file →
Q4_K_S20.4 GB 21.6 GB
Too big Get file →
Q4_K_M21.7 GB 22.9 GB
Too big Get file →
Q4_122.4 GB 23.6 GB
Too big Get file →
Q5_K_S24.6 GB 25.8 GB
Too big Get file →
Q5_K_M25.3 GB 26.5 GB
Too big Get file →
Q6_K29.2 GB 30.4 GB
Too big Get file →

For 8 GB VRAM, use Q4_K_S (size 20.37 GB, ~21.6 GB memory). For 12 GB VRAM, Q4_K_M (size 21.71 GB, ~22.9 GB memory) is recommended. For 16 GB VRAM, Q5_K_S (size 24.56 GB, ~25.8 GB memory) provides a good balance of quality and memory usage. Estimates are approximate.

What GPU do you need for KAT Coder V2.5 Dev 35B A3B MTP ABLITERATED?

These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.

  • 4 GB VRAM: the best quant that fits is the main file (0.2 GB file, about 1.4 GB in use).
  • 6 GB VRAM: the best quant that fits is the main file (0.2 GB file, about 1.4 GB in use).
  • 8 GB VRAM: the best quant that fits is the main file (0.2 GB file, about 1.4 GB in use).
  • 12 GB VRAM: the best quant that fits is the main file (0.2 GB file, about 1.4 GB in use).
  • 16 GB VRAM: the best quant that fits is Q2_K (13.2 GB file, about 14.4 GB in use).
  • 24 GB VRAM: the best quant that fits is Q4_1 (22.4 GB file, about 23.6 GB in use).
  • 32 GB VRAM: the best quant that fits is Q6_K (29.2 GB file, about 30.4 GB in use).

Which KAT Coder V2.5 Dev 35B A3B MTP ABLITERATED quant should you choose?

Q4_K_M is the usual starting point for KAT Coder V2.5 Dev 35B A3B MTP ABLITERATED: it balances file size and quality well. Here it is a 21.7 GB file.

If memory is tight, the smallest file is the first one in the table at 0.2 GB, with the quality trade-offs that come with fewer bits.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architectureqwen35moe
Parameters35.5B (approximate, mixture of experts)
Active parameters3B
Layers41
Trained context262,144 tokens
KV cache (fp16)82 KB per token

Frequently asked questions about KAT Coder V2.5 Dev 35B A3B MTP ABLITERATED

How much VRAM does KAT Coder V2.5 Dev 35B A3B MTP ABLITERATED need?

About 22.9 GB at Q4_K_M with an 8K context window. That is the 21.7 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.

Can I run KAT Coder V2.5 Dev 35B A3B MTP ABLITERATED on an 8 GB GPU?

Yes, with the smaller files: the 0.2 GB file needs about 1.4 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of KAT Coder V2.5 Dev 35B A3B MTP ABLITERATED GGUF?

It depends on the quant: file is 0.2 GB, Q2_K is 13.2 GB, Q3_K_S is 15.5 GB, IQ3_S is 15.6 GB, IQ3_M is 15.8 GB, Q3_K_M is 17.2 GB and 9 more.

What context length does KAT Coder V2.5 Dev 35B A3B MTP ABLITERATED support?

The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 0.7 GB at 8K and 2.8 GB at 32K.

Where can I download KAT Coder V2.5 Dev 35B A3B MTP ABLITERATED GGUF?

The files are on Hugging Face in the mradermacher/KAT-Coder-V2.5-Dev-35B-A3B-MTP-ABLITERATED-i1-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does KAT Coder V2.5 Dev 35B A3B MTP ABLITERATED use?

Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.

Can I run this model on 8 GB VRAM?

Use the Q4_K_S quant, which requires about 21.6 GB of VRAM at 8k context. 8 GB VRAM is insufficient for full precision.

Does this model support vision?

No, the model does not include vision capabilities.

What license applies to this model?

The model is released under the Apache‑2.0 license.

Other GGUFs of KAT-Coder-V2.5-Dev-35B-A3B-MTP-ABLITERATED

Similar models

Browse all →

More from mradermacher

All by mradermacher →

More Kat models

All Kat →

Other large models

All large models →