⚡ 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.

Qwen3.6 35B-A3B — Claude 4.7 Opus Reasoning Distilled — APEX GGUF

APEX (Adaptive Precision for EXpert Models) quantizations of lordx64/Qwen3.6-35B-A3B-Claude-4.7-Opus-Reasoning-Distilled.

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

Available Files

File	Profile	Size	Best For
Qwen3.6-35B-A3B-Claude-4.7-Opus-Reasoning-Distilled-APEX-I-Balanced.gguf	I-Balanced	24 GB	Best overall quality/size ratio
Qwen3.6-35B-A3B-Claude-4.7-Opus-Reasoning-Distilled-APEX-Balanced.gguf	Balanced	24 GB	General purpose
Qwen3.6-35B-A3B-Claude-4.7-Opus-Reasoning-Distilled-APEX-I-Quality.gguf	I-Quality	21 GB	Highest quality with imatrix
Qwen3.6-35B-A3B-Claude-4.7-Opus-Reasoning-Distilled-APEX-Quality.gguf	Quality	21 GB	Highest quality standard
Qwen3.6-35B-A3B-Claude-4.7-Opus-Reasoning-Distilled-APEX-I-Compact.gguf	I-Compact	16 GB	Consumer GPUs, best quality/size
Qwen3.6-35B-A3B-Claude-4.7-Opus-Reasoning-Distilled-APEX-Compact.gguf	Compact	16 GB	Consumer GPUs
Qwen3.6-35B-A3B-Claude-4.7-Opus-Reasoning-Distilled-APEX-I-Mini.gguf	I-Mini	13 GB	Smallest "safe" tier
Qwen3.6-35B-A3B-Claude-4.7-Opus-Reasoning-Distilled-APEX-I-Nano.gguf	I-Nano	11 GB	Experimental — IQ2_XXS mid-layer experts
Qwen3.6-35B-A3B-Claude-4.7-Opus-Reasoning-Distilled-F16.gguf	F16 reference	65 GB	Full-precision reference
mmproj.gguf	Vision projector	~1 GB	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 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 (first/last 5) and keeping attention, SSM/Mamba, and shared expert tensors at higher precision.

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

Nano (experimental tier)

The APEX Nano tier pushes mid-layer routed experts to IQ2_XXS (2.06 bpw), near-edge to IQ2_S, edges to Q3_K, with shared experts kept at Q5_K. About 20% smaller than Mini with modest quality cost — viable only on MoE thanks to sparse per-token expert activation. Requires imatrix.

Benchmarks pending. Feedback welcome.

Architecture

Model: Qwen3.6 35B-A3B Claude 4.7 Opus Reasoning Distilled
Base: Qwen 3.6 35B-A3B
Layers: 40
Experts: 256 routed + shared (8 active per token)
Total Parameters: ~35B
Active Parameters: ~3B per token
Attention: Hybrid (full attention every 4th layer, linear/Mamba 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, Wikipedia)

Run with LocalAI

local-ai run mudler/Qwen3.6-35B-A3B-Claude-4.7-Opus-Reasoning-Distilled-APEX-GGUF@Qwen3.6-35B-A3B-Claude-4.7-Opus-Reasoning-Distilled-APEX-I-Balanced.gguf