⚡ 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
Nex-N2-mini — APEX GGUF
APEX (Adaptive Precision for EXpert Models) quantizations of nex-agi/Nex-N2-mini — an agentic model with Agentic Thinking, post-trained on Qwen3.5-35B-A3B-Base for first-tier coding and long-horizon agentic tasks.
Brought to you by the LocalAI team | APEX Project | Technical Report
Available Files
| File | Profile | Best For |
|---|---|---|
| Nex-N2-mini-APEX-I-Balanced.gguf | I-Balanced | Best overall — imatrix-enhanced, lowest worst-case divergence |
| Nex-N2-mini-APEX-I-Quality.gguf | I-Quality | Highest quality with imatrix |
| Nex-N2-mini-APEX-Quality.gguf | Quality | Highest quality (no imatrix) |
| Nex-N2-mini-APEX-Balanced.gguf | Balanced | General purpose |
| Nex-N2-mini-APEX-I-Compact.gguf | I-Compact | Consumer GPUs, imatrix-enhanced |
| Nex-N2-mini-APEX-Compact.gguf | Compact | Consumer GPUs |
| Nex-N2-mini-APEX-I-Mini.gguf | I-Mini | Smallest viable, fastest inference |
| mmproj.gguf | Vision projector | Required for image understanding |
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 get more aggressive compression. I-variants use diverse imatrix calibration (chat, code, reasoning, tool-calling, agentic traces, Wikipedia).
The key insight: in MoE models, expert FFN tensors make up the bulk of model weight but only ~8/256 experts activate per token. APEX compresses middle-layer experts more aggressively while preserving edge layers and keeping attention, SSM/Mamba, and shared-expert tensors at higher precision.
See the APEX project for full details, technical report, and scripts.
Architecture
- Model: Nex-N2-mini (Qwen3_5MoeForConditionalGeneration, post-trained on Qwen3.5-35B-A3B-Base)
- Layers: 40
- Experts: 256 routed + 1 shared (8 active per token)
- Total Parameters: ~35B
- Active Parameters: ~3B per token
- Attention: Hybrid (full attention every 4th layer, linear otherwise)
- Vision: Built-in vision encoder (mmproj included)
- APEX Config: 5+5 symmetric edge gradient across 40 layers
- Calibration: v1.3 diverse dataset (chat, code, reasoning, multilingual, tool-calling, agentic traces, Wikipedia)
Run with LocalAI
local-ai run mudler/Nex-N2-mini-APEX-GGUF@Nex-N2-mini-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. Base model by Nex-AGI.