{
  "id": "llamacpp-spark-q4km",
  "title": "llama.cpp on Spark: Q4_K_M prefill and decode",
  "kind": "benchmark",
  "source_type": "community_operator_report",
  "summary": "Original Spark report separating prompt processing from single-batch generation at different context depths.",
  "runtime": "llama.cpp",
  "runtime_topics": [
    "llama.cpp"
  ],
  "hardware_topics": [
    "spark",
    "nvidia"
  ],
  "platform_topics": [
    "linux"
  ],
  "model_label": "Qwen3 / Ministral 3",
  "model_family": "Multiple",
  "model_id": null,
  "model_release_date": null,
  "model_artifact": null,
  "evidence_state": "source_reported",
  "site_reproduced": false,
  "record_type": "curated_reference",
  "performance_ranking_allowed": false,
  "metrics": null,
  "metrics_status": "site_metrics_unmeasured; attributed_source_figures_in_editorial.reported_results",
  "conditions": {
    "model_revision": null,
    "weight_hash": null,
    "runtime_execution_revision": null,
    "input_tokens": null,
    "output_tokens": null,
    "concurrency": null,
    "cache_state": null,
    "quantization": "Q4_K_M",
    "speculative_decoding": null,
    "temperature": null,
    "top_p": null
  },
  "reported_setup": {
    "system": "NVIDIA DGX Spark / GB10 / 128 GB LPDDR5x",
    "node_count": 1,
    "parallelism": null,
    "kv_dtype": null,
    "context_limit_tokens": null,
    "experiment_date": null
  },
  "test_profiles": [
    {
      "label": "Prompt processing",
      "evidence_url": "https://github.com/DandinPower/llama.cpp_bench/blob/a1e9514561e9be68982e1905f48808143082013a/dgx_spark/report.md",
      "conditions": {
        "model_revision": null,
        "weight_hash": null,
        "runtime_execution_revision": null,
        "input_tokens": null,
        "output_tokens": null,
        "concurrency": null,
        "cache_state": null,
        "quantization": "Q4_K_M",
        "speculative_decoding": null,
        "temperature": null,
        "top_p": null,
        "context_depth_tokens": [
          512,
          16384
        ]
      }
    },
    {
      "label": "Generation",
      "evidence_url": "https://github.com/DandinPower/llama.cpp_bench/blob/a1e9514561e9be68982e1905f48808143082013a/dgx_spark/report.md",
      "conditions": {
        "model_revision": null,
        "weight_hash": null,
        "runtime_execution_revision": null,
        "input_tokens": null,
        "output_tokens": null,
        "concurrency": null,
        "cache_state": null,
        "quantization": "Q4_K_M",
        "speculative_decoding": null,
        "temperature": null,
        "top_p": null,
        "batch_size": 1,
        "context_depth_tokens": [
          512,
          2048
        ]
      }
    }
  ],
  "deployment_options": [],
  "recipe_parameters": null,
  "measurement_scope": "inference_speed",
  "notes": [
    "Q4_K_M is the selected weight format.",
    "Prefill and decode tables describe different context depths."
  ],
  "limitations": [
    "Exact model and runtime commits are not specified in this report.",
    "A tested context depth is not the machine’s maximum supported context."
  ],
  "comparison_status": "not_comparable",
  "editorial_status": "source_derived_detailed_article",
  "public_benchmark_snapshot": false,
  "source": {
    "publisher": "DandinPower",
    "url": "https://github.com/DandinPower/llama.cpp_bench",
    "repository": "DandinPower/llama.cpp_bench",
    "platform": "github",
    "revision": "a1e9514561e9be68982e1905f48808143082013a",
    "document_path": "dgx_spark/report.md",
    "pinned_url": "https://github.com/DandinPower/llama.cpp_bench/blob/a1e9514561e9be68982e1905f48808143082013a/dgx_spark/report.md",
    "report_url": "https://github.com/DandinPower/llama.cpp_bench/blob/a1e9514561e9be68982e1905f48808143082013a/dgx_spark/report.md",
    "document_sha256": "44d415d26c12533d1694fcc7b622f8961ea0a320078177e14c663a5b77028c70",
    "last_commit_at": "2025-12-05T13:40:35Z",
    "checked_on": "2026-10-17",
    "metadata_status": "checked"
  },
  "license": {
    "declared_spdx": null,
    "scope": "Repository metadata only; per-file, model, data and conversion terms remain separate."
  },
  "editorial": {
    "version": 1,
    "published_on": "2026-10-17",
    "modified_on": "2026-10-17",
    "byline": "Run AI Here editorial",
    "creation_method": "AI-assisted source reading and original editorial synthesis; no independent model execution.",
    "answer": "The report studies Qwen3 and Ministral 3 Q4_K_M artifacts on a single GB10 Spark with 128 GB unified memory. Prompt processing is tabulated at context depths 512 and 16,384; single-batch generation at 512 and 2,048. Context depth is the source table’s label, not a specified generated-output budget or a maximum capacity test.",
    "sections": [
      {
        "heading": "What experiment is being described?",
        "paragraphs": [
          "The report studies Qwen3 and Ministral 3 Q4_K_M artifacts on a single GB10 Spark with 128 GB unified memory. Prompt processing is tabulated at context depths 512 and 16,384; single-batch generation at 512 and 2,048. Context depth is the source table’s label, not a specified generated-output budget or a maximum capacity test."
        ],
        "citations": [
          "https://github.com/DandinPower/llama.cpp_bench/blob/a1e9514561e9be68982e1905f48808143082013a/dgx_spark/report.md#L20"
        ]
      },
      {
        "heading": "What is a useful concrete comparison?",
        "paragraphs": [
          "The selected Qwen3 30B MoE row reports 89.3 decode tokens/s at depth 512 and 83.8 at 2,048, while the 32B dense row reports 10.7 at depth 512. These results illustrate the author’s bandwidth interpretation for this particular pair of Q4_K_M artifacts. They do not establish equal quality or the same outcome for another MoE family."
        ],
        "citations": [
          "https://github.com/DandinPower/llama.cpp_bench/blob/a1e9514561e9be68982e1905f48808143082013a/dgx_spark/report.md#L33"
        ]
      },
      {
        "heading": "What prevents a strong reproduction claim?",
        "paragraphs": [
          "The report does not identify exact weight hashes, executed runtime commit, output lengths, cache policy or repetitions. Those gaps remain explicit. Without them, a later reader cannot know whether a discrepancy comes from kernels, artifacts or workload. A tested context depth also cannot certify the machine’s maximum context."
        ],
        "citations": [
          "https://github.com/DandinPower/llama.cpp_bench/blob/a1e9514561e9be68982e1905f48808143082013a/dgx_spark/report.md#L18"
        ]
      },
      {
        "heading": "How is the source reused?",
        "paragraphs": [
          "This article gives a small attributed factual selection with an original explanation. Repository metadata does not declare a license, so the full result table and prose are left with the original author. No independent run is claimed, and prefill and decode remain separate measurement types."
        ],
        "citations": [
          "https://github.com/DandinPower/llama.cpp_bench/blob/a1e9514561e9be68982e1905f48808143082013a/dgx_spark/report.md#L46"
        ]
      }
    ],
    "facts": [],
    "reported_results": [
      {
        "label": "Qwen3 30B MoE, shallow context",
        "metric": "decode",
        "value": 89.3,
        "unit": "tokens/s",
        "statistic": "reported value",
        "uncertainty": null,
        "conditions": {
          "system": "NVIDIA DGX Spark / GB10 / 128 GB LPDDR5x",
          "node_count": 1,
          "parallelism": null,
          "kv_dtype": null,
          "context_limit_tokens": null,
          "experiment_date": null,
          "runtime": "llama.cpp, executed commit unspecified",
          "quantization": "Q4_K_M",
          "batch_size": 1,
          "input_tokens": null,
          "output_tokens": null,
          "repetitions": null,
          "cache_state": null,
          "model": "Qwen3 30B MoE",
          "context_depth_tokens": 512
        },
        "evidence_state": "source_reported",
        "site_reproduced": false,
        "evidence_url": "https://github.com/DandinPower/llama.cpp_bench/blob/a1e9514561e9be68982e1905f48808143082013a/dgx_spark/report.md#L40",
        "source_document_sha256": "44d415d26c12533d1694fcc7b622f8961ea0a320078177e14c663a5b77028c70"
      },
      {
        "label": "Qwen3 30B MoE, deeper context",
        "metric": "decode",
        "value": 83.8,
        "unit": "tokens/s",
        "statistic": "reported value",
        "uncertainty": null,
        "conditions": {
          "system": "NVIDIA DGX Spark / GB10 / 128 GB LPDDR5x",
          "node_count": 1,
          "parallelism": null,
          "kv_dtype": null,
          "context_limit_tokens": null,
          "experiment_date": null,
          "runtime": "llama.cpp, executed commit unspecified",
          "quantization": "Q4_K_M",
          "batch_size": 1,
          "input_tokens": null,
          "output_tokens": null,
          "repetitions": null,
          "cache_state": null,
          "model": "Qwen3 30B MoE",
          "context_depth_tokens": 2048
        },
        "evidence_state": "source_reported",
        "site_reproduced": false,
        "evidence_url": "https://github.com/DandinPower/llama.cpp_bench/blob/a1e9514561e9be68982e1905f48808143082013a/dgx_spark/report.md#L40",
        "source_document_sha256": "44d415d26c12533d1694fcc7b622f8961ea0a320078177e14c663a5b77028c70"
      },
      {
        "label": "Qwen3 32B dense, shallow context",
        "metric": "decode",
        "value": 10.7,
        "unit": "tokens/s",
        "statistic": "reported value",
        "uncertainty": null,
        "conditions": {
          "system": "NVIDIA DGX Spark / GB10 / 128 GB LPDDR5x",
          "node_count": 1,
          "parallelism": null,
          "kv_dtype": null,
          "context_limit_tokens": null,
          "experiment_date": null,
          "runtime": "llama.cpp, executed commit unspecified",
          "quantization": "Q4_K_M",
          "batch_size": 1,
          "input_tokens": null,
          "output_tokens": null,
          "repetitions": null,
          "cache_state": null,
          "model": "Qwen3 32B dense",
          "context_depth_tokens": 512
        },
        "evidence_state": "source_reported",
        "site_reproduced": false,
        "evidence_url": "https://github.com/DandinPower/llama.cpp_bench/blob/a1e9514561e9be68982e1905f48808143082013a/dgx_spark/report.md#L41",
        "source_document_sha256": "44d415d26c12533d1694fcc7b622f8961ea0a320078177e14c663a5b77028c70"
      }
    ],
    "source_document_sha256": "44d415d26c12533d1694fcc7b622f8961ea0a320078177e14c663a5b77028c70"
  }
}