⚡ Each donation = another big MoE quantized

I host 30+ free APEX MoE quantizations as independent research. My only local hardware is an NVIDIA DGX Spark (122 GB unified memory), enough for ~30-50B-class MoEs, but bigger ones (200B+) require rented compute on H100/H200/Blackwell, typically $20-100 per quant.
If APEX quants are useful to you, your support directly funds those bigger runs.

🎉 Patreon (Monthly)  |  ☕ Buy Me a Coffee  |  ⭐ GitHub Sponsors

KAT-Coder-V2.5-Dev — APEX GGUF

APEX (Adaptive Precision for EXpert Models) quantizations of Kwaipilot/KAT-Coder-V2.5-Dev — Kwaipilot's Mixture-of-Experts model for agentic coding.

Brought to you by the LocalAI team | APEX Project | Technical Report

Available Files

What is APEX?

APEX is a quantization strategy for Mixture-of-Experts (MoE) models. It classifies tensors by role (routed expert, shared expert, attention) and applies a layer-wise precision gradient — edge layers (first/last 5) get higher precision, middle layers compress more aggressively. I-variants use diverse imatrix calibration (chat, code, reasoning, tool-calling, agentic traces, Wikipedia).

In MoE models the routed-expert FFN tensors dominate the weight budget but only ~8/256 experts activate per token, so APEX compresses middle-layer experts hardest while preserving edge layers, attention, and the always-active shared expert.

See the APEX project for full details.

Architecture

• Model: KAT-Coder-V2.5-Dev (Qwen3_5MoeForConditionalGeneration)
• Layers: 40 · Experts: 256 routed + 1 shared (8 active per token)
• Attention: 16 heads / 2 KV, hybrid (full attention every 4th layer)
• Calibration: v1.3 diverse dataset

Note: the config advertises an image token, but the released checkpoint ships no vision encoder weights, so these are text-only GGUFs (no mmproj).

Run with LocalAI

local-ai run mudler/KAT-Coder-V2.5-Dev-APEX-GGUF@KAT-Coder-V2.5-Dev-APEX-I-Balanced.gguf

Credits

APEX is brought to you by the LocalAI team. Built on llama.cpp. Base model by Kwaipilot.