{
  "model": "glm-5.2",
  "endpoint": "http://<head-node>:8210",
  "config": {
    "max_model_len": 128000,
    "mtp": false,
    "async_scheduling": false,
    "enforce_eager": true,
    "kv_cache_dtype": "fp8_ds_mla"
  },
  "concurrency": [
    {
      "n": 1,
      "wall_s": 83.46870014700107,
      "agg_prefill_tps": 23730.91664685037,
      "agg_decode_tps": 12.319039106857042,
      "total_in": 10086,
      "total_out": 1024,
      "per": [
        {
          "ttft_s": 0.4250151879969053,
          "duration_s": 83.46720631397329,
          "in_tokens": 10086,
          "out_tokens": 1024,
          "prefill_tps": 23730.91664685037,
          "decode_tps": 12.319039106857042
        }
      ]
    },
    {
      "n": 2,
      "wall_s": 87.5483156919945,
      "agg_prefill_tps": 23861.97882361139,
      "agg_decode_tps": 23.45067654874787,
      "total_in": 20172,
      "total_out": 2048,
      "per": [
        {
          "ttft_s": 0.8448660710128024,
          "duration_s": 87.54601570201339,
          "in_tokens": 10086,
          "out_tokens": 1024,
          "prefill_tps": 11937.9867958352,
          "decode_tps": 11.799151503225504
        },
        {
          "ttft_s": 0.29956138995476067,
          "duration_s": 87.36476089898497,
          "in_tokens": 10086,
          "out_tokens": 1024,
          "prefill_tps": 33669.225535117104,
          "decode_tps": 11.749815147370066
        }
      ]
    },
    {
      "n": 4,
      "wall_s": 105.01616257702699,
      "agg_prefill_tps": 43985.403645281614,
      "agg_decode_tps": 39.081311826937394,
      "total_in": 40344,
      "total_out": 4096,
      "per": [
        {
          "ttft_s": 0.9164160549989901,
          "duration_s": 105.01215200102888,
          "in_tokens": 10086,
          "out_tokens": 1024,
          "prefill_tps": 11005.9180488835,
          "decode_tps": 9.827491882380185
        },
        {
          "ttft_s": 0.30762922001304105,
          "duration_s": 104.68880904599791,
          "in_tokens": 10086,
          "out_tokens": 1024,
          "prefill_tps": 32786.22232170414,
          "decode_tps": 9.800617330686007
        },
        {
          "ttft_s": 0.915373956959229,
          "duration_s": 105.011793638987,
          "in_tokens": 10086,
          "out_tokens": 1024,
          "prefill_tps": 11018.44762276674,
          "decode_tps": 9.827427332513922
        },
        {
          "ttft_s": 0.9172133629908785,
          "duration_s": 105.01317776099313,
          "in_tokens": 10086,
          "out_tokens": 1024,
          "prefill_tps": 10996.350911320404,
          "decode_tps": 9.827470314686213
        }
      ]
    },
    {
      "n": 8,
      "wall_s": 211.22031849296764,
      "agg_prefill_tps": 667.7085918195853,
      "agg_decode_tps": 38.802276633220906,
      "total_in": 80688,
      "total_out": 8192,
      "per": [
        {
          "ttft_s": 1.0292239459813572,
          "duration_s": 120.49851802602643,
          "in_tokens": 10086,
          "out_tokens": 1024,
          "prefill_tps": 9799.616535722043,
          "decode_tps": 8.562869713740708
        },
        {
          "ttft_s": 1.0304710670025088,
          "duration_s": 120.49879872903693,
          "in_tokens": 10086,
          "out_tokens": 1024,
          "prefill_tps": 9787.756612457557,
          "decode_tps": 8.562938981568225
        },
        {
          "ttft_s": 1.0277425730018876,
          "duration_s": 120.49542883899994,
          "in_tokens": 10086,
          "out_tokens": 1024,
          "prefill_tps": 9813.74155839458,
          "decode_tps": 8.562984954126112
        },
        {
          "ttft_s": 1.0274803399806842,
          "duration_s": 120.49637578701368,
          "in_tokens": 10086,
          "out_tokens": 1024,
          "prefill_tps": 9816.246216632824,
          "decode_tps": 8.56289828555041
        },
        {
          "ttft_s": 120.84070815495215,
          "duration_s": 211.21325369097758,
          "in_tokens": 10086,
          "out_tokens": 1024,
          "prefill_tps": 83.46525069239813,
          "decode_tps": 11.319809505556076
        },
        {
          "ttft_s": 0.298364263959229,
          "duration_s": 119.94888777396409,
          "in_tokens": 10086,
          "out_tokens": 1024,
          "prefill_tps": 33804.31646257152,
          "decode_tps": 8.549899908415021
        },
        {
          "ttft_s": 1.0294912179815583,
          "duration_s": 120.49788634100696,
          "in_tokens": 10086,
          "out_tokens": 1024,
          "prefill_tps": 9797.072402205451,
          "decode_tps": 8.562934146278115
        },
        {
          "ttft_s": 120.49913692096015,
          "duration_s": 211.05247776099714,
          "in_tokens": 10086,
          "out_tokens": 1024,
          "prefill_tps": 83.70184432620277,
          "decode_tps": 11.297208810960775
        }
      ]
    },
    {
      "n": 16,
      "wall_s": 347.18223452800885,
      "agg_prefill_tps": 649.7799636587389,
      "agg_decode_tps": 47.188063449126105,
      "total_in": 161376,
      "total_out": 16384,
      "per": [
        {
          "ttft_s": 122.9887021080358,
          "duration_s": 247.88026835903293,
          "in_tokens": 10086,
          "out_tokens": 1024,
          "prefill_tps": 82.00753261986821,
          "decode_tps": 8.19110553825593
        },
        {
          "ttft_s": 122.36689803999616,
          "duration_s": 247.4718603939982,
          "in_tokens": 10086,
          "out_tokens": 1024,
          "prefill_tps": 82.42425166897135,
          "decode_tps": 8.177133670407716
        },
        {
          "ttft_s": 247.88314656500006,
          "duration_s": 347.00379909202456,
          "in_tokens": 10086,
          "out_tokens": 1024,
          "prefill_tps": 40.688526589100896,
          "decode_tps": 10.320755300930719
        },
        {
          "ttft_s": 122.98994975897949,
          "duration_s": 247.8806138219661,
          "in_tokens": 10086,
          "out_tokens": 1024,
          "prefill_tps": 82.00670070819035,
          "decode_tps": 8.191164709349822
        },
        {
          "ttft_s": 122.98767868900904,
          "duration_s": 247.87981558503816,
          "in_tokens": 10086,
          "out_tokens": 1024,
          "prefill_tps": 82.00821503025367,
          "decode_tps": 8.191068112251395
        },
        {
          "ttft_s": 248.3539364010212,
          "duration_s": 347.174491420039,
          "in_tokens": 10086,
          "out_tokens": 1024,
          "prefill_tps": 40.61139576106404,
          "decode_tps": 10.352097291936138
        },
        {
          "ttft_s": 0.3048746809945442,
          "duration_s": 121.96259336097864,
          "in_tokens": 10086,
          "out_tokens": 1024,
          "prefill_tps": 33082.445439870724,
          "decode_tps": 8.408837606851414
        },
        {
          "ttft_s": 0.9096906300401315,
          "duration_s": 122.36533842602512,
          "in_tokens": 10086,
          "out_tokens": 1024,
          "prefill_tps": 11087.285794682804,
          "decode_tps": 8.422827744646204
        },
        {
          "ttft_s": 122.99076496699126,
          "duration_s": 247.88210159598384,
          "in_tokens": 10086,
          "out_tokens": 1024,
          "prefill_tps": 82.00615715096104,
          "decode_tps": 8.19112059821224
        },
        {
          "ttft_s": 0.9164128599804826,
          "duration_s": 122.37234718800755,
          "in_tokens": 10086,
          "out_tokens": 1024,
          "prefill_tps": 11005.956420357095,
          "decode_tps": 8.422807873982435
        },
        {
          "ttft_s": 122.98619927599793,
          "duration_s": 247.8764023249969,
          "in_tokens": 10086,
          "out_tokens": 1024,
          "prefill_tps": 82.00920151508731,
          "decode_tps": 8.191194945840868
        },
        {
          "ttft_s": 0.9112317039980553,
          "duration_s": 122.36756711598719,
          "in_tokens": 10086,
          "out_tokens": 1024,
          "prefill_tps": 11068.534990329446,
          "decode_tps": 8.422780059433755
        },
        {
          "ttft_s": 248.34920644800877,
          "duration_s": 347.16969710600097,
          "in_tokens": 10086,
          "out_tokens": 1024,
          "prefill_tps": 40.612169228378335,
          "decode_tps": 10.35210403417749
        },
        {
          "ttft_s": 0.9128333679982461,
          "duration_s": 122.36809424101375,
          "in_tokens": 10086,
          "out_tokens": 1024,
          "prefill_tps": 11049.114059138315,
          "decode_tps": 8.422854577452778
        },
        {
          "ttft_s": 0.909883382031694,
          "duration_s": 122.3648867320153,
          "in_tokens": 10086,
          "out_tokens": 1024,
          "prefill_tps": 11084.937036083458,
          "decode_tps": 8.422872436569227
        },
        {
          "ttft_s": 248.3545818660059,
          "duration_s": 347.175478707999,
          "in_tokens": 10086,
          "out_tokens": 1024,
          "prefill_tps": 40.61129021344842,
          "decode_tps": 10.35206148387519
        }
      ]
    }
  ],
  "niah": {
    "in_tokens": 126579,
    "ttft_s": 232.2551459139795,
    "duration_s": 232.2551459139795,
    "out_tokens": 1,
    "prefill_tps": 544.9997652447329,
    "decode_tps": 0,
    "found": false,
    "needle_value": "BANANA-MOOSE-7421",
    "answer_head": ""
  }
}
