lovedheart/qwen3-next-reap-25b-a3b-instruct-gguf MXFP4_MOE GGUF - Free GGUF Download is indexed on GraySoft with repository links, GGUF quant files, and Hugging Face metadata. This page helps you pick a local model for guIDE or other runtimes. See related models in the same shard below.
Model Intelligence Sheet
lovedheart/qwen3-next-reap-25b-a3b-instruct-gguf overview
!qwen3-next-instruction Qwen3-Next-REAP-25B-A3B-Instruct has the following specifications: Number of Linear Attention Heads: 32 for V and 16 for QK Head Dimension: 128 Live test video: https://www.bilibili.com/video/BV19AkABKEdG/?vd_source=448090107c928cea02cdf07046d02784
Downloads
86
Likes
2
Pipeline
—
Library
—
Visibility
Public
Access
Open
Repository Files & Downloads
4 files detected
Direct downloads for all repository files
Model Details Live
Metadata Inspector
Normalized metadata (stored in metadata_json)
{
"metadata": {},
"card_data": {
"base_model": [
"Qwen/Qwen3-Next-80B-A3B-Instruct"
],
"tags": [
"text-generation-inference"
],
"license": "apache-2.0",
"frontmatter": {
"base_model": [
"Qwen/Qwen3-Next-80B-A3B-Instruct"
],
"tags": [
"text-generation-inference"
],
"license": "apache-2.0"
},
"hero_image_url": "https://cdn-uploads.huggingface.co/production/uploads/68121d80da035a609e569a81/Ft9cmZlll_PehtFYkESxH.png",
"summary": "!qwen3-next-instruction **Qwen3-Next-REAP-25B-A3B-Instruct** has the following specifications: **Number of Linear Attention Heads: 32 for V and 16 for QK **Head Dimension: 128 Live test video: https://www.bilibili.com/video/BV19AkABKEdG/?vd_source=448090107c928cea02cdf07046d02784",
"quick_links": [],
"benchmark_table_html": "",
"readme_markdown": "---\nbase_model:\n- Qwen/Qwen3-Next-80B-A3B-Instruct\ntags:\n- text-generation-inference\nlicense: apache-2.0\n---\n\n\n\n\n\n**Qwen3-Next-REAP-25B-A3B-Instruct** has the following specifications:\n\n- **Type:** Causal Language Models\n- **Number of Parameters**: 25B in total and 3B activated\n- **Hidden Dimension**: 2048\n- **Number of Layers**: 48\n- **Hybrid Layout**: 12 * (3 * (Gated DeltaNet -> MoE) -> 1 * (Gated Attention -> MoE))\n- **Gated Attention**:\n- **Number of Attention Heads**: 16 for Q and 2 for KV\n- **Head Dimension**: 256\n- **Rotary Position Embedding Dimension**: 64\n- **Gated DeltaNet**: \n **Number of Linear Attention Heads: 32 for V and 16 for QK \n **Head Dimension: 128\n- **Mixture of Experts**:\n- **Number of Experts: 160 (uniformly pruned from 512)\n- **Number of Activated Experts: 10\n- **Number of Shared Experts: 1\n- **Context Length**: 262,144 natively and extensible up to 1,010,000 tokens\n- **Compression Method**: REAP (Router-weighted Expert Activation Pruning)\n- **Compression Ratio**: 68.75% expert pruning\n- **Specialized**: Coding and Agent\n\nLive test video: https://www.bilibili.com/video/BV19AkABKEdG/?vd_source=448090107c928cea02cdf07046d02784",
"related_quantizations": []
},
"tags": [
"gguf",
"text-generation-inference",
"base_model:Qwen/Qwen3-Next-80B-A3B-Instruct",
"base_model:quantized:Qwen/Qwen3-Next-80B-A3B-Instruct",
"license:apache-2.0",
"endpoints_compatible",
"region:us",
"conversational"
],
"likes": 2,
"downloads": 86,
"gated": false,
"private": false,
"last_modified": "2026-01-22T21:51:21.000Z",
"created_at": "2026-01-15T15:45:46.000Z",
"pipeline_tag": "",
"library_name": ""
}
Source payload excerpt (from Hugging Face API)
{
"_id": "69690baa213899f567089204",
"id": "lovedheart/Qwen3-Next-REAP-25B-A3B-Instruct-GGUF",
"modelId": "lovedheart/Qwen3-Next-REAP-25B-A3B-Instruct-GGUF",
"sha": "12bf71a7b4c21d31ecb2273f950cb3168791334a",
"createdAt": "2026-01-15T15:45:46.000Z",
"lastModified": "2026-01-22T21:51:21.000Z",
"author": "lovedheart",
"downloads": 86,
"likes": 2,
"gated": false,
"private": false,
"pipeline_tag": "",
"library_name": "",
"siblings_count": 6
}