dinerburger/qwen3.5-35b-a3b-gguf overview

{
  "metadata": {},
  "card_data": {
    "license": "apache-2.0",
    "base_model": [
      "Qwen/Qwen3.5-35B-A3B"
    ],
    "frontmatter": {
      "license": "apache-2.0",
      "base_model": [
        "Qwen/Qwen3.5-35B-A3B"
      ]
    },
    "hero_image_url": "",
    "summary": "This is an IQ4_NL quantization of Qwen3.5-35B-A3B, using the unsloth imatrix data, but with the following special rules applied: The full quantization script is here: `` QUANT=\"IQ4_NL\" llama-quantize \\ --output-tensor-type bf16 \\ --token-embedding-type bf16 \\ --tensor-type attn_qkv=bf16 \\ --tensor-type attn_v=bf16 \\ --tensor-type attn_q=bf16 \\ --tensor-type attn_k=bf16 \\ --tensor-type attn_gate=bf16 \\ --tensor-type ssm_ba=bf16 \\ --tensor-type ssm_beta=bf16 \\ --tensor-type ssm_alpha=bf16 \\ --tensor-type ssm_out=bf16 \\ --tensor-type ffn_down_shexp=bf16 \\ --tensor-type ffn_gate_shexp=bf16 \\ --tensor-type ffn_up_shexp=bf16 \\ --imatrix Qwen3.5-35B-A3B-imatrix.gguf_file \\ Qwen3.5-35B-A3B.bf16.gguf \\ Qwen3.5-35B-A3B.${QUANT}.gguf \\ ${QUANT} ``",
    "quick_links": [],
    "benchmark_table_html": "",
    "readme_markdown": "---\nlicense: apache-2.0\nbase_model:\n- Qwen/Qwen3.5-35B-A3B\n---\n\n\nThis is an IQ4_NL quantization of [Qwen3.5-35B-A3B](https://huggingface.co/Qwen/Qwen3.5-35B-A3B), using the [unsloth imatrix data](https://huggingface.co/unsloth/Qwen3.5-35B-A3B-GGUF/resolve/main/imatrix_unsloth.gguf_file), but with the following special rules applied:\n\n- The embedding and output layers were kept in BF16\n- All SSM tensors were left in BF16\n- All attention tensors were left in BF16\n- Shared expert tensors were left in BF16\n- All other tensors use IQ4_NL\n\nThe full quantization script is here:\n```\nQUANT=\"IQ4_NL\"\nllama-quantize \\\n  --output-tensor-type bf16 \\\n  --token-embedding-type bf16 \\\n  --tensor-type attn_qkv=bf16 \\\n  --tensor-type attn_v=bf16 \\\n  --tensor-type attn_q=bf16 \\\n  --tensor-type attn_k=bf16 \\\n  --tensor-type attn_gate=bf16 \\\n  --tensor-type ssm_ba=bf16 \\\n  --tensor-type ssm_beta=bf16 \\\n  --tensor-type ssm_alpha=bf16 \\\n  --tensor-type ssm_out=bf16 \\\n  --tensor-type ffn_down_shexp=bf16 \\\n  --tensor-type ffn_gate_shexp=bf16 \\\n  --tensor-type ffn_up_shexp=bf16 \\\n  --imatrix Qwen3.5-35B-A3B-imatrix.gguf_file \\\n  Qwen3.5-35B-A3B.bf16.gguf \\\n  Qwen3.5-35B-A3B.${QUANT}.gguf \\\n  ${QUANT}\n\n```",
    "related_quantizations": []
  },
  "tags": [
    "gguf",
    "base_model:Qwen/Qwen3.5-35B-A3B",
    "base_model:quantized:Qwen/Qwen3.5-35B-A3B",
    "license:apache-2.0",
    "endpoints_compatible",
    "region:us",
    "imatrix",
    "conversational"
  ],
  "likes": 0,
  "downloads": 212,
  "gated": false,
  "private": false,
  "last_modified": "2026-02-27T18:27:08.000Z",
  "created_at": "2026-02-27T17:45:40.000Z",
  "pipeline_tag": "",
  "library_name": ""
}

{
  "_id": "69a1d844b83b7eeb978351d7",
  "id": "dinerburger/Qwen3.5-35B-A3B-GGUF",
  "modelId": "dinerburger/Qwen3.5-35B-A3B-GGUF",
  "sha": "67172544d956a12f2ac70867cf99ac35a82e6cbf",
  "createdAt": "2026-02-27T17:45:40.000Z",
  "lastModified": "2026-02-27T18:27:08.000Z",
  "author": "dinerburger",
  "downloads": 212,
  "likes": 0,
  "gated": false,
  "private": false,
  "pipeline_tag": "",
  "library_name": "",
  "siblings_count": 3
}