{
  "arch_id": "qwen3-next-mtp",
  "artifact_role": "forge-local",
  "base_trunk": "/Users/youssof/.mtplx/models/Qwen--Qwen3.8-27B",
  "exactness_baseline": {},
  "forge_provenance": {
    "depth_default_note": "Depth pin REMOVED 2026-08-14 after the gated live ABBA: D2 39.5 vs D3 40.6 tok/s matched-window (tie; D3 fewer verify rounds 697 vs 883). The earlier forge-verify \"D3 18.8 collapse\" was order/JIT-confounded (mistakes/single-forge-verify-tune-rows...). Family ceiling D3 applies.",
    "forge_inputs": {
      "mtp_source_path": "/Users/youssof/.mtplx/models/Qwen--Qwen3.8-27B",
      "trunk_path": "/Users/youssof/.mtplx/models/Qwen3.8-27B-MTPLX-Optimized-Quality"
    },
    "forge_recipe": {
      "body_bits": 8,
      "body_group_size": 64,
      "body_mode": "affine",
      "mtp_policy": "keep_bf16"
    },
    "forged_at": "2026-08-14T11:23:31-07:00",
    "forged_locally": true,
    "mtp_contract": {
      "base_hidden_variant": "post_norm",
      "concat_order": "embedding_hidden",
      "hidden_variant": "post_norm",
      "mtp_position_mode": "local",
      "mtp_quant_group_size": 64,
      "mtp_quant_mode": "affine"
    },
    "mtplx_version": "2.6.0",
    "published_to_hf": null,
    "recommended_profile_note": "Restamped sustained->turbo 2026-08-14: the sustained recommendation came from the same confounded verify rows; every gated arm served turbo, medium Flappy 59.8 tok/s blended quiet-window.",
    "source_format": "bf16_native",
    "source_repo": "/Users/youssof/.mtplx/models/Qwen--Qwen3.8-27B",
    "source_sha": null,
    "head_quantization": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine",
      "policy": "all",
      "quantized_at": "2026-08-20T04:50:00Z",
      "source_sidecar_bytes": 849400388,
      "quantized_sidecar_bytes": 451270880,
      "tool": "scripts/build_qwen38_q4head_sidecar.py",
      "note": "Structural head quantization of the released sidecar; no calibration, no training. Trunk weights unchanged."
    }
  },
  "mtp_contract": {
    "base_hidden_variant": "post_norm",
    "concat_order": "embedding_hidden",
    "hidden_variant": "post_norm",
    "mtp_position_mode": "local",
    "mtp_quant_group_size": 64,
    "mtp_quant_mode": "affine"
  },
  "mtp_depth_default": 3,
  "mtp_depth_max": 3,
  "mtp_sidecar": "int8-g64-prequantized",
  "mtplx_version": "2.9.0",
  "precision_policy": {
    "intended_default_for": [
      "m1",
      "m2"
    ],
    "note": "FP16 precision sibling of Qwen3.8-27B-MTPLX-Optimized-Quality: byte-identical INT packs, bf16 floats cast to fp16 for M1/M2 Macs. Same tuned depth/draft-sampler stamps as the parent; not a separate speed claim.",
    "routing": "mtplx start auto-selects this artifact on M1/M2 Apple Silicon",
    "variant": "fp16"
  },
  "precision_variant": "fp16",
  "public_model_id": "mtplx-qwen38-27b-optimized-quality-fp16",
  "recommended_draft_sampler": {
    "temperature": 1.0,
    "top_k": 20,
    "top_p": 0.95
  },
  "recommended_profile": "turbo",
  "release_validation": {
    "draft_sampler_transfer_note": {
      "basis": "dropday uncapped ABBA campaign (clean-receipts.jsonl, temp 1.0 xhigh/medium, die-gated, max fans verified)",
      "note": "draft 1.0 adopted by transfer: OS uncapped ABBA (+13%) + Bare scaffold strict A/B (keep_official_draft_sampler_1_0); same MTP head architecture and temp-1.0 target contract. OQ depth arms measured at draft 0.6; app-receipt validates live at ship config.",
      "stamped_at": "2026-08-14T21:45:00-07:00"
    },
    "dropday_depth_abba": {
      "basis": "dropday uncapped ABBA campaign (clean-receipts.jsonl, temp 1.0 xhigh/medium, die-gated, max fans verified)",
      "decode_tps_blended_d2": [
        27.3,
        28.0
      ],
      "decode_tps_blended_d3": [
        33.2,
        33.1
      ],
      "stamped_at": "2026-08-14T21:45:00-07:00",
      "verdict": "depth_3 (+19.9% blended)",
      "workload": "simple flappy xhigh uncapped (33k-46k tokens, EOS stop)"
    }
  },
  "sampler": {
    "temperature": 1.0,
    "top_k": 20,
    "top_p": 0.95
  },
  "speed_evidence": {
    "depth": 3,
    "tok_s": [
      48.714441349701545
    ],
    "acceptance_by_depth": [
      0.9838709677419355,
      0.946236559139785,
      0.8870967741935484
    ],
    "greedy_diagnostic": {
      "tok_s": 17.324087175386293
    },
    "forge_verify_rows": [
      {
        "acceptance_by_position": [],
        "depth": 0,
        "finish_reasons": {
          "stop": 1
        },
        "hit_token_budget": false,
        "hit_token_budget_count": 0,
        "multiplier_vs_ar": 1.0,
        "quality_passed": true,
        "tok_s": 17.324087175386293,
        "verify_time_s": 77.89033425005618
      },
      {
        "acceptance_by_position": [
          0.9744637385086823
        ],
        "depth": 1,
        "finish_reasons": {
          "stop": 1
        },
        "hit_token_budget": false,
        "hit_token_budget_count": 0,
        "multiplier_vs_ar": 1.5802488858523818,
        "quality_passed": true,
        "tok_s": 27.376369457313725,
        "verify_time_s": 71.90131354611367
      },
      {
        "acceptance_by_position": [
          0.9586466165413534,
          0.8909774436090225
        ],
        "depth": 2,
        "finish_reasons": {
          "stop": 1
        },
        "hit_token_budget": false,
        "hit_token_budget_count": 0,
        "multiplier_vs_ar": 2.3137630346497304,
        "quality_passed": true,
        "tok_s": 40.08383251545826,
        "verify_time_s": 20.266882561845705
      },
      {
        "acceptance_by_position": [
          0.9838709677419355,
          0.946236559139785,
          0.8870967741935484
        ],
        "depth": 3,
        "finish_reasons": {
          "stop": 1
        },
        "hit_token_budget": false,
        "hit_token_budget_count": 0,
        "multiplier_vs_ar": 2.811948523262687,
        "quality_passed": true,
        "tok_s": 48.714441349701545,
        "verify_time_s": 12.932098778896034
      }
    ],
    "quality_rejected": [],
    "acceptance_collapsed": [],
    "failure_reasons": [],
    "verdict": "mtp_depth_wins",
    "artifact_fingerprint": "sha256:ff1082fa404d8a02077c776f377d4ac9e4d151f0caf427a8baeda4b63905c05f"
  },
  "verified_on": {
    "timestamp": "2026-08-19T23:23:22-07:00",
    "hardware": "macOS-26.3.1-arm64-arm-64bit-Mach-O",
    "machine_arch": "arm64",
    "macos": "26.3.1",
    "model": "Qwen3.8-27B-MTPLX-Optimized-Quality-FP16"
  }
}