{
  "id": "gpt-oss-20b-vllm",
  "title": "GPT-OSS 20B: vLLM recipe",
  "kind": "deployment",
  "source_type": "maintainer_recipe",
  "summary": "MXFP4 serving recipe with accelerator allowlists and model-specific serving controls.",
  "runtime": "vLLM",
  "runtime_topics": [
    "vLLM"
  ],
  "hardware_topics": [
    "amd",
    "cpu",
    "intel",
    "nvidia"
  ],
  "platform_topics": [
    "linux"
  ],
  "model_label": "gpt-oss-20b",
  "model_family": "GPT-OSS",
  "model_id": "openai/gpt-oss-20b",
  "model_release_date": null,
  "model_artifact": {
    "id": "openai/gpt-oss-20b",
    "revision": null,
    "format": "mxfp4"
  },
  "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": null,
    "speculative_decoding": null,
    "temperature": null,
    "top_p": null
  },
  "reported_setup": {
    "system": null,
    "node_count": null,
    "parallelism": null,
    "kv_dtype": null,
    "context_limit_tokens": null,
    "experiment_date": null
  },
  "test_profiles": [],
  "deployment_options": [
    {
      "variant": "default",
      "model_id": "openai/gpt-oss-20b",
      "precision": "mxfp4",
      "supported_hardware": [
        "h100",
        "h200",
        "b200",
        "mi300x",
        "mi325x",
        "mi355x",
        "arc_pro_b60",
        "arc_pro_b70",
        "xeon6"
      ],
      "model_revision": null,
      "declared_planning_minimum_gb": 16,
      "minimum_runtime_version": "0.10.1",
      "extra_args": [],
      "tensor_parallel_by_hardware": null
    }
  ],
  "recipe_parameters": {
    "minimum_runtime_version": "0.10.1",
    "nightly_required": false,
    "model_context_limit_tokens": 131072,
    "compatible_strategies": [
      "single_node_tp"
    ],
    "hardware_claims": [
      {
        "profile": "h100",
        "maintainer_status": "verified"
      },
      {
        "profile": "h200",
        "maintainer_status": "verified"
      },
      {
        "profile": "b200",
        "maintainer_status": "verified"
      },
      {
        "profile": "mi300x",
        "maintainer_status": "verified"
      },
      {
        "profile": "mi325x",
        "maintainer_status": "verified"
      },
      {
        "profile": "mi355x",
        "maintainer_status": "verified"
      },
      {
        "profile": "arc_pro_b60",
        "maintainer_status": "verified"
      },
      {
        "profile": "arc_pro_b70",
        "maintainer_status": "verified"
      },
      {
        "profile": "xeon6",
        "maintainer_status": "verified"
      }
    ],
    "recipe_updated_on": "2026-09-12",
    "effective_default_minimum_runtime_version": "0.10.1"
  },
  "measurement_scope": "deployment_reference",
  "notes": [
    "Recipe minimum engine version: 0.10.1.",
    "Default artifact: openai/gpt-oss-20b.",
    "Variant and hardware settings are source declarations; select the matching original profile."
  ],
  "limitations": [
    "Use the exact engine and artifact instructions; this entry is not a fresh model release.",
    "Input length, output policy, concurrency and cache are workload-specific; no single benchmark is inferred from the recipe."
  ],
  "comparison_status": "not_comparable",
  "editorial_status": "source_derived_detailed_article",
  "public_benchmark_snapshot": false,
  "source": {
    "publisher": "vllm-project",
    "url": "https://github.com/vllm-project/recipes",
    "repository": "vllm-project/recipes",
    "platform": "github",
    "revision": "d5fc1f55cee70f4e1944f2dd4f7569cb8276ad94",
    "document_path": "models/openai/gpt-oss-20b.yaml",
    "pinned_url": "https://github.com/vllm-project/recipes/blob/d5fc1f55cee70f4e1944f2dd4f7569cb8276ad94/models/openai/gpt-oss-20b.yaml",
    "report_url": null,
    "document_sha256": "0cb6342ac0c6225518d4a62677b75aa952e03cca2ffe44df1b836f284c373a46",
    "last_commit_at": "2026-10-17T09:22:45Z",
    "checked_on": "2026-10-17",
    "metadata_status": "checked"
  },
  "license": {
    "declared_spdx": "Apache-2.0",
    "scope": "Recipe repository only; each model and conversion has independent distribution terms."
  },
  "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 pinned gpt-oss-20b recipe defaults to openai/gpt-oss-20b (mxfp4). Its effective default-variant minimum is vLLM 0.10.1. Select the exact variant, hardware override and topology before treating it as a launch plan. No deployment was run by this site.",
    "sections": [
      {
        "heading": "Which artifact and engine does this recipe select?",
        "paragraphs": [
          "The recipe identifies openai/gpt-oss-20b as its model family entry and selects openai/gpt-oss-20b by default. The default precision label is mxfp4; model-wide and variant-specific runtime requirements are kept separately below. The document revision fixes what the recipe said, while the weight revision and executed runtime revision remain separate fields. A model-family title alone cannot establish either artifact identity."
        ],
        "citations": [
          "https://github.com/vllm-project/recipes/blob/d5fc1f55cee70f4e1944f2dd4f7569cb8276ad94/models/openai/gpt-oss-20b.yaml#L52"
        ]
      },
      {
        "heading": "What needs special attention for this model?",
        "paragraphs": [
          "The model name says 20b while the recipe summarizes approximately 21B total and 3.6B active parameters. Its MXFP4 expert representation and explicit accelerator allowlist must stay with the artifact. The declared 16 GB planning minimum does not establish the maximum context or user count of a consumer GPU."
        ],
        "citations": [
          "https://github.com/vllm-project/recipes/blob/d5fc1f55cee70f4e1944f2dd4f7569cb8276ad94/models/openai/gpt-oss-20b.yaml#L25"
        ]
      },
      {
        "heading": "How should the settings be applied?",
        "paragraphs": [
          "Choose one artifact variant and one hardware profile from the original recipe, then resolve its shared, variant, hardware, strategy and optional-feature settings together. The base arguments listed here are only one layer and are deliberately not presented as a complete runnable command. Check any nightly-image requirement and the model’s parsers before loading weights. A context declaration is a model limit, while batched-token and sequence settings control a serving workload."
        ],
        "citations": [
          "https://github.com/vllm-project/recipes/blob/d5fc1f55cee70f4e1944f2dd4f7569cb8276ad94/models/openai/gpt-oss-20b.yaml#L59"
        ]
      },
      {
        "heading": "What does the memory number mean?",
        "paragraphs": [
          "The GB figure is the recipe’s declared planning minimum for the selected variant. It is not a site measurement, a universal peak allocation or a promise of maximum context. KV cache, auxiliary/draft models, image encoders, temporary workspace and active requests still require memory. Keep GB and GiB distinct, and use the actual artifact size and runtime allocation for a machine-specific budget."
        ],
        "citations": [
          "https://github.com/vllm-project/recipes/blob/d5fc1f55cee70f4e1944f2dd4f7569cb8276ad94/models/openai/gpt-oss-20b.yaml#L55"
        ]
      },
      {
        "heading": "What is needed to reproduce a deployment?",
        "paragraphs": [
          "Save the model and conversion revisions, weight hashes, tokenizer/template, container or runtime digest, device count/topology and complete resolved launch. First check one short public request. For a subsequent measurement, fix input and output counts, concurrency, cache state, warmup, sampling, reasoning effort, repetitions and failures. The declarations on this page are source configuration facts; no throughput or quality outcome follows from them."
        ],
        "citations": [
          "https://github.com/vllm-project/recipes/blob/d5fc1f55cee70f4e1944f2dd4f7569cb8276ad94/models/openai/gpt-oss-20b.yaml#L88"
        ]
      }
    ],
    "facts": [
      {
        "name": "Architecture",
        "value": "moe",
        "classification": "recipe_declaration",
        "evidence_url": "https://github.com/vllm-project/recipes/blob/d5fc1f55cee70f4e1944f2dd4f7569cb8276ad94/models/openai/gpt-oss-20b.yaml#L28",
        "source_document_sha256": "0cb6342ac0c6225518d4a62677b75aa952e03cca2ffe44df1b836f284c373a46"
      },
      {
        "name": "Total parameters as counted by recipe",
        "value": "21B",
        "classification": "recipe_declaration",
        "evidence_url": "https://github.com/vllm-project/recipes/blob/d5fc1f55cee70f4e1944f2dd4f7569cb8276ad94/models/openai/gpt-oss-20b.yaml#L29",
        "source_document_sha256": "0cb6342ac0c6225518d4a62677b75aa952e03cca2ffe44df1b836f284c373a46"
      },
      {
        "name": "Active parameters as counted by recipe",
        "value": "3.6B",
        "classification": "recipe_declaration",
        "evidence_url": "https://github.com/vllm-project/recipes/blob/d5fc1f55cee70f4e1944f2dd4f7569cb8276ad94/models/openai/gpt-oss-20b.yaml#L30",
        "source_document_sha256": "0cb6342ac0c6225518d4a62677b75aa952e03cca2ffe44df1b836f284c373a46"
      },
      {
        "name": "Native context declaration (tokens)",
        "value": 131072,
        "classification": "recipe_declaration",
        "evidence_url": "https://github.com/vllm-project/recipes/blob/d5fc1f55cee70f4e1944f2dd4f7569cb8276ad94/models/openai/gpt-oss-20b.yaml#L31",
        "source_document_sha256": "0cb6342ac0c6225518d4a62677b75aa952e03cca2ffe44df1b836f284c373a46"
      },
      {
        "name": "Model-wide minimum vLLM",
        "value": "0.10.1",
        "classification": "recipe_declaration",
        "evidence_url": "https://github.com/vllm-project/recipes/blob/d5fc1f55cee70f4e1944f2dd4f7569cb8276ad94/models/openai/gpt-oss-20b.yaml#L27",
        "source_document_sha256": "0cb6342ac0c6225518d4a62677b75aa952e03cca2ffe44df1b836f284c373a46"
      },
      {
        "name": "Effective default-variant minimum vLLM",
        "value": "0.10.1",
        "classification": "recipe_declaration",
        "evidence_url": "https://github.com/vllm-project/recipes/blob/d5fc1f55cee70f4e1944f2dd4f7569cb8276ad94/models/openai/gpt-oss-20b.yaml#L27",
        "source_document_sha256": "0cb6342ac0c6225518d4a62677b75aa952e03cca2ffe44df1b836f284c373a46"
      },
      {
        "name": "Default artifact",
        "value": "openai/gpt-oss-20b",
        "classification": "recipe_declaration",
        "evidence_url": "https://github.com/vllm-project/recipes/blob/d5fc1f55cee70f4e1944f2dd4f7569cb8276ad94/models/openai/gpt-oss-20b.yaml#L52",
        "source_document_sha256": "0cb6342ac0c6225518d4a62677b75aa952e03cca2ffe44df1b836f284c373a46"
      },
      {
        "name": "Default precision label",
        "value": "mxfp4",
        "classification": "recipe_declaration",
        "evidence_url": "https://github.com/vllm-project/recipes/blob/d5fc1f55cee70f4e1944f2dd4f7569cb8276ad94/models/openai/gpt-oss-20b.yaml#L52",
        "source_document_sha256": "0cb6342ac0c6225518d4a62677b75aa952e03cca2ffe44df1b836f284c373a46"
      },
      {
        "name": "Default planning minimum (GB; recipe declaration)",
        "value": 16,
        "classification": "recipe_declaration",
        "evidence_url": "https://github.com/vllm-project/recipes/blob/d5fc1f55cee70f4e1944f2dd4f7569cb8276ad94/models/openai/gpt-oss-20b.yaml#L55",
        "source_document_sha256": "0cb6342ac0c6225518d4a62677b75aa952e03cca2ffe44df1b836f284c373a46"
      },
      {
        "name": "Default variant hardware allowlist",
        "value": [
          "h100",
          "h200",
          "b200",
          "mi300x",
          "mi325x",
          "mi355x",
          "arc_pro_b60",
          "arc_pro_b70",
          "xeon6"
        ],
        "classification": "recipe_declaration",
        "evidence_url": "https://github.com/vllm-project/recipes/blob/d5fc1f55cee70f4e1944f2dd4f7569cb8276ad94/models/openai/gpt-oss-20b.yaml#L52",
        "source_document_sha256": "0cb6342ac0c6225518d4a62677b75aa952e03cca2ffe44df1b836f284c373a46"
      },
      {
        "name": "Shared base arguments (incomplete launch)",
        "value": [],
        "classification": "recipe_declaration",
        "evidence_url": "https://github.com/vllm-project/recipes/blob/d5fc1f55cee70f4e1944f2dd4f7569cb8276ad94/models/openai/gpt-oss-20b.yaml#L52",
        "source_document_sha256": "0cb6342ac0c6225518d4a62677b75aa952e03cca2ffe44df1b836f284c373a46"
      },
      {
        "name": "Declared compatible strategies",
        "value": [
          "single_node_tp"
        ],
        "classification": "recipe_declaration",
        "evidence_url": "https://github.com/vllm-project/recipes/blob/d5fc1f55cee70f4e1944f2dd4f7569cb8276ad94/models/openai/gpt-oss-20b.yaml#L52",
        "source_document_sha256": "0cb6342ac0c6225518d4a62677b75aa952e03cca2ffe44df1b836f284c373a46"
      },
      {
        "name": "Optional feature names",
        "value": [
          "tool_calling",
          "spec_decoding"
        ],
        "classification": "recipe_declaration",
        "evidence_url": "https://github.com/vllm-project/recipes/blob/d5fc1f55cee70f4e1944f2dd4f7569cb8276ad94/models/openai/gpt-oss-20b.yaml#L52",
        "source_document_sha256": "0cb6342ac0c6225518d4a62677b75aa952e03cca2ffe44df1b836f284c373a46"
      },
      {
        "name": "Features selected by default",
        "value": [
          "spec_decoding"
        ],
        "classification": "recipe_declaration",
        "evidence_url": "https://github.com/vllm-project/recipes/blob/d5fc1f55cee70f4e1944f2dd4f7569cb8276ad94/models/openai/gpt-oss-20b.yaml#L52",
        "source_document_sha256": "0cb6342ac0c6225518d4a62677b75aa952e03cca2ffe44df1b836f284c373a46"
      }
    ],
    "reported_results": [],
    "source_document_sha256": "0cb6342ac0c6225518d4a62677b75aa952e03cca2ffe44df1b836f284c373a46"
  }
}