---
base_model: inclusionAI/Ling-3.0-flash
library_name: transformers
tags: [reap, expert-pruning, moe, bailingmoe]
---

**This is an experimental REAP.**

# Ling-3.0-flash REAP384 (97B total / 5.1B active)

[384 of 512 routed experts kept per layer - 25% of experts pruned]
from [inclusionAI/Ling-3.0-flash](https://huggingface.co/inclusionAI/Ling-3.0-flash)
(124B total / 5.1B active).

**Method:** one-shot REAP ([Router-weighted Expert Activation Pruning](https://arxiv.org/abs/2510.13999)) -
experts scored by router-gate-value × output-L2-norm over calibration data, lowest-scoring deleted.
No fine-tuning, no recovery training.

**Calibration:** 1M tokens of ultrachat (chat-only calibration)

BF16 safetensors. Loads with `trust_remote_code=True` (custom `bailing_hybrid` / BailingMoeV3 code).
Research artifact - quantized builds live in the sibling `-GGUF` repo.

2026-08-21: chat_template.jinja updated to support `reasoning_effort` (low = no thinking, high = thinking, default unchanged).