aessedai/minimax-m2.5-gguf overview

{
  "metadata": {},
  "card_data": {
    "base_model": [
      "MiniMaxAI/MiniMax-M2.5"
    ],
    "frontmatter": {
      "base_model": [
        "MiniMaxAI/MiniMax-M2.5"
      ]
    },
    "hero_image_url": "kld_data/01_kld_vs_filesize_pareto.png \"Chart showing Pareto KLD analysis of quants\"",
    "summary": "This repo contains specialized MoE-quants for MiniMax-M2.5. The idea being that given the huge size of the FFN tensors compared to the rest of the tensors in the model, it should be possible to achieve a better quality while keeping the overall size of the entire model smaller compared to a similar naive quantization. To that end, the quantization type default is kept in high quality and the FFN UP + FFN GATE tensors are quanted down along with the FFN DOWN tensors. | Quant | Size | Mixture | PPL | 1-(Mean PPL(Q)/PPL(base)) | KLD | | :--------- | :--------- | :------- | :------- | :------- | :------- | | Q5_K_M | 157.23 GiB (5.91 BPW) | Q8_0 / Q5_K / Q5_K / Q6_K | 7.126261 ± 0.115850 | +0.5877% | 0.023465 ± 0.001079 | | Q4_K_M | 130.52 GiB (4.90 BPW) | Q8_0 / Q4_K / Q4_K / Q5_K| 7.173459 ± 0.116673 | +1.2462% | 0.041269 ± 0.001426 | IQ4_XS | 101.10 GiB (3.80 BPW) | Q8_0 / IQ3_S / IQ3_S / IQ4_XS | 7.513587 ± 0.122746 | +6.0549% | 0.095077 ± 0.002168 | | IQ3_S | 78.76 GiB (2.96 BPW) | Q8_0 / IQ2_S / IQ2_S / IQ3_S | 8.284882 ± 0.135705 | +16.9418% | 0.244096 ± 0.004148 | Provided here as well as a couple of graphs showing the Pareto frontier for KLD and PPL for my quants vs Unsloth. Full graphs of all of the quants are available in the kld_data directory, as well as the raw data broken down per quant as well as a CSV with the collated data. While the PPL between the quant methods is similar, I feel like the KLD of the quants provided here are slightly better and that these quants will offer better long context performance due to keeping the default type as Q8_0. This comes with a slight performance penalty in PP / TG due to the higher quality quantization but I think the tradeoff is worthwhile. !kld_graph !ppl_graph",
    "quick_links": [],
    "benchmark_table_html": "",
    "readme_markdown": "---\nbase_model:\n- MiniMaxAI/MiniMax-M2.5\n---\n\nThis repo contains specialized MoE-quants for MiniMax-M2.5. The idea being that given the huge size of the FFN tensors compared to the rest of the tensors in the model, it should be possible to achieve a better quality while keeping the overall size of the entire model smaller compared to a similar naive quantization. To that end, the quantization type default is kept in high quality and the FFN UP + FFN GATE tensors are quanted down along with the FFN DOWN tensors.\n\n| Quant | Size | Mixture | PPL | 1-(Mean PPL(Q)/PPL(base)) | KLD | \n| :--------- | :--------- | :------- | :------- | :------- | :------- |\n| Q5_K_M | 157.23 GiB (5.91 BPW) | Q8_0 / Q5_K / Q5_K / Q6_K | 7.126261 ± 0.115850 | +0.5877% | 0.023465 ± 0.001079 |\n| Q4_K_M | 130.52 GiB (4.90 BPW) | Q8_0 / Q4_K / Q4_K / Q5_K| 7.173459 ± 0.116673 | +1.2462% | 0.041269 ± 0.001426\n| IQ4_XS | 101.10 GiB (3.80 BPW) | Q8_0 / IQ3_S / IQ3_S / IQ4_XS | 7.513587 ± 0.122746 | +6.0549% | 0.095077 ± 0.002168 |\n| IQ3_S | 78.76 GiB (2.96 BPW) | Q8_0 / IQ2_S / IQ2_S / IQ3_S | 8.284882 ± 0.135705 | +16.9418% | 0.244096 ± 0.004148 |\n\nProvided here as well as a couple of graphs showing the Pareto frontier for KLD and PPL for my quants vs Unsloth. \n\nFull graphs of all of the quants are available in the `kld_data` directory, as well as the raw data broken down per quant as well as a CSV with the collated data.\n\nWhile the PPL between the quant methods is similar, I feel like the KLD of the quants provided here are slightly better and that these quants will offer better long context performance due to keeping the default type as Q8_0. This comes with a slight performance penalty in PP / TG due to the higher quality quantization but I think the tradeoff is worthwhile.\n\n![kld_graph](kld_data/01_kld_vs_filesize_pareto.png \"Chart showing Pareto KLD analysis of quants\")\n![ppl_graph](kld_data/02_ppl_vs_filesize_pareto.png \"Chart showing Pareto PPL analysis of quants\")",
    "related_quantizations": []
  },
  "tags": [
    "gguf",
    "base_model:MiniMaxAI/MiniMax-M2.5",
    "base_model:quantized:MiniMaxAI/MiniMax-M2.5",
    "endpoints_compatible",
    "region:us",
    "imatrix",
    "conversational"
  ],
  "likes": 27,
  "downloads": 1102,
  "gated": false,
  "private": false,
  "last_modified": "2026-02-16T03:02:47.000Z",
  "created_at": "2026-02-14T01:26:30.000Z",
  "pipeline_tag": "",
  "library_name": ""
}

{
  "_id": "698fcf4661b2cd203fc29f65",
  "id": "AesSedai/MiniMax-M2.5-GGUF",
  "modelId": "AesSedai/MiniMax-M2.5-GGUF",
  "sha": "1e997fe11a8a1726a938fe9679ded03eda7285ca",
  "createdAt": "2026-02-14T01:26:30.000Z",
  "lastModified": "2026-02-16T03:02:47.000Z",
  "author": "AesSedai",
  "downloads": 1102,
  "likes": 27,
  "gated": false,
  "private": false,
  "pipeline_tag": "",
  "library_name": "",
  "siblings_count": 55
}