mudler/MiniMax-M2.5-APEX-GGUF

⚡ Each donation = another big MoE quantized

I host 25+ 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

💚 Big thanks to Hugging Face for generously donating additional storage — much appreciated.

MiniMax-M2.5 APEX GGUF

APEX (Adaptive Precision for EXpert Models) quantizations of MiniMax-M2.5.

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

Benchmark Results

Benchmarks coming soon. For reference APEX benchmarks on the Qwen3.5-35B-A3B architecture, see mudler/Qwen3.5-35B-A3B-APEX-GGUF.

Available Files

File	Profile	Size	Best For
MiniMax-M2.5-APEX-I-Balanced.gguf	I-Balanced	155 GB	Best overall quality/size ratio
MiniMax-M2.5-APEX-I-Quality.gguf	I-Quality	130 GB	Highest quality with imatrix
MiniMax-M2.5-APEX-Quality.gguf	Quality	130 GB	Highest quality standard
MiniMax-M2.5-APEX-Balanced.gguf	Balanced	155 GB	General purpose
MiniMax-M2.5-APEX-I-Compact.gguf	I-Compact	100 GB	Multi-GPU setups, best quality/size
MiniMax-M2.5-APEX-Compact.gguf	Compact	100 GB	Multi-GPU setups
MiniMax-M2.5-APEX-I-Mini.gguf	I-Mini	81 GB	Smallest viable

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 get higher precision, middle layers get more aggressive compression. I-variants use diverse imatrix calibration (chat, code, reasoning, tool-calling, agentic traces, Wikipedia).

See the APEX project for full details, technical report, and scripts.

Architecture

• Model: MiniMax-M2.5 (MiniMaxM2)
• Layers: 62
• Experts: 256 routed + 1 shared (8 active per token)
• Total Parameters: 228.7B
• Active Parameters: ~45B per token
• APEX Config: 5+5 symmetric edge gradient across 62 layers
• Calibration: v1.3 diverse dataset (chat, code, reasoning, multilingual, tool-calling, Wikipedia)

Run with LocalAI

local-ai run mudler/MiniMax-M2.5-APEX-GGUF@MiniMax-M2.5-APEX-I-Balanced.gguf

Credits

APEX is brought to you by the LocalAI team. Developed through human-driven, AI-assisted research. Built on llama.cpp.