Pelican MapAN AI FIELD GUIDE
HF / OPENENV · REFERENCE ONLY

nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4

Upstream outputs, not a Pelican Map ranking

Upstream outputs, not a Pelican Map ranking

An independent HuggingFace / OpenEnv scoring reference, excluded from the main timeline, collection and case total.

Rows: 20 · Viewable images: 20 · Upstream mean reward: 0.814219

Species words in upstream blind captions

bird: 7/20 · duck: 1/20 · goose: 1/20 · not identified: 2/20 · pelican: 7/20 · stork: 2/20

Matches species words in original captions; no new image judging. Multiple labels and missing/unidentified captions remain in the denominator. Original captions are retained below.

nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__001
Perfect upstream reward = 1

Upstream outputs, not a Pelican Map ranking · reward 1.000

A cartoon pelican is riding a bicycle on a line under a blue sky with a cloud.

Upstream SVG ↗ · Local SVG

Original fields and blind caption
{
  "model": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4",
  "access": "open",
  "provider": "hf-inference-providers",
  "task": "pelican_bicycle",
  "sample": 1,
  "reward": 1.0,
  "structure_score": 1.0,
  "semantic_score": 1.0,
  "gate_passed": true,
  "violations": [],
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "A cartoon pelican is riding a bicycle on a line under a blue sky with a cloud.",
  "latency": 31.8,
  "svg": "svg/nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__001.svg",
  "png": "png/nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__001.png"
}
nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__000
Sample not called a pelican by the blind caption

Upstream outputs, not a Pelican Map ranking · reward 0.810

A bird with an orange beak is riding a bicycle with two wheels and handlebars.

Upstream SVG ↗ · Local SVG

Original fields and blind caption
{
  "model": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4",
  "access": "open",
  "provider": "hf-inference-providers",
  "task": "pelican_bicycle",
  "sample": 0,
  "reward": 0.8104166666666666,
  "structure_score": 1.0,
  "semantic_score": 0.7083333333333333,
  "gate_passed": true,
  "violations": [],
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "A bird with an orange beak is riding a bicycle with two wheels and handlebars.",
  "latency": 268.67,
  "svg": "svg/nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__000.svg",
  "png": "png/nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__000.png"
}
Expand all samples (20)
nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__000
nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__000

Upstream outputs, not a Pelican Map ranking · reward 0.810

A bird with an orange beak is riding a bicycle with two wheels and handlebars.

Upstream SVG ↗ · Local SVG

Original fields and blind caption
{
  "model": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4",
  "access": "open",
  "provider": "hf-inference-providers",
  "task": "pelican_bicycle",
  "sample": 0,
  "reward": 0.8104166666666666,
  "structure_score": 1.0,
  "semantic_score": 0.7083333333333333,
  "gate_passed": true,
  "violations": [],
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "A bird with an orange beak is riding a bicycle with two wheels and handlebars.",
  "latency": 268.67,
  "svg": "svg/nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__000.svg",
  "png": "png/nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__000.png"
}
nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__001
nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__001

Upstream outputs, not a Pelican Map ranking · reward 1.000

A cartoon pelican is riding a bicycle on a line under a blue sky with a cloud.

Upstream SVG ↗ · Local SVG

Original fields and blind caption
{
  "model": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4",
  "access": "open",
  "provider": "hf-inference-providers",
  "task": "pelican_bicycle",
  "sample": 1,
  "reward": 1.0,
  "structure_score": 1.0,
  "semantic_score": 1.0,
  "gate_passed": true,
  "violations": [],
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "A cartoon pelican is riding a bicycle on a line under a blue sky with a cloud.",
  "latency": 31.8,
  "svg": "svg/nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__001.svg",
  "png": "png/nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__001.png"
}
nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__002
nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__002

Upstream outputs, not a Pelican Map ranking · reward 0.838

A cartoon goose is riding a red bicycle.

Upstream SVG ↗ · Local SVG

Original fields and blind caption
{
  "model": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4",
  "access": "open",
  "provider": "hf-inference-providers",
  "task": "pelican_bicycle",
  "sample": 2,
  "reward": 0.8375,
  "structure_score": 1.0,
  "semantic_score": 0.75,
  "gate_passed": true,
  "violations": [],
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "A cartoon goose is riding a red bicycle.",
  "latency": 68.4,
  "svg": "svg/nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__002.svg",
  "png": "png/nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__002.png"
}
nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__003
nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__003

Upstream outputs, not a Pelican Map ranking · reward 0.465

A cartoon bird with an orange beak and feet is riding a bicycle with two large wheels.

Upstream SVG ↗ · Local SVG

Original fields and blind caption
{
  "model": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4",
  "access": "open",
  "provider": "hf-inference-providers",
  "task": "pelican_bicycle",
  "sample": 3,
  "reward": 0.46510416666666665,
  "structure_score": 1.0,
  "semantic_score": 0.17708333333333331,
  "gate_passed": true,
  "violations": [],
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "A cartoon bird with an orange beak and feet is riding a bicycle with two large wheels.",
  "latency": 35.8,
  "svg": "svg/nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__003.svg",
  "png": "png/nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__003.png"
}
nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__004
nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__004

Upstream outputs, not a Pelican Map ranking · reward 0.847

The image shows a cartoon-style drawing of a white bird with an orange beak riding a bicycle.

Upstream SVG ↗ · Local SVG

Original fields and blind caption
{
  "model": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4",
  "access": "open",
  "provider": "hf-inference-providers",
  "task": "pelican_bicycle",
  "sample": 4,
  "reward": 0.8465277777777778,
  "structure_score": 1.0,
  "semantic_score": 0.7638888888888888,
  "gate_passed": true,
  "violations": [],
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "The image shows a cartoon-style drawing of a white bird with an orange beak riding a bicycle.",
  "latency": 69.25,
  "svg": "svg/nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__004.svg",
  "png": "png/nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__004.png"
}
nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__005
nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__005

Upstream outputs, not a Pelican Map ranking · reward 0.765

A stick figure with an orange beak is riding a bicycle.

Upstream SVG ↗ · Local SVG

Original fields and blind caption
{
  "model": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4",
  "access": "open",
  "provider": "hf-inference-providers",
  "task": "pelican_bicycle",
  "sample": 5,
  "reward": 0.7652777777777777,
  "structure_score": 1.0,
  "semantic_score": 0.6388888888888888,
  "gate_passed": true,
  "violations": [],
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "A stick figure with an orange beak is riding a bicycle.",
  "latency": 28.18,
  "svg": "svg/nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__005.svg",
  "png": "png/nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__005.png"
}
nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__006
nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__006

Upstream outputs, not a Pelican Map ranking · reward 0.765

A cartoon stork is riding a bicycle under a sunny sky with clouds.

Upstream SVG ↗ · Local SVG

Original fields and blind caption
{
  "model": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4",
  "access": "open",
  "provider": "hf-inference-providers",
  "task": "pelican_bicycle",
  "sample": 6,
  "reward": 0.7652777777777777,
  "structure_score": 1.0,
  "semantic_score": 0.6388888888888888,
  "gate_passed": true,
  "violations": [],
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "A cartoon stork is riding a bicycle under a sunny sky with clouds.",
  "latency": 24.75,
  "svg": "svg/nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__006.svg",
  "png": "png/nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__006.png"
}
nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__007
nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__007

Upstream outputs, not a Pelican Map ranking · reward 0.847

A bird with an orange beak is riding a bicycle.

Upstream SVG ↗ · Local SVG

Original fields and blind caption
{
  "model": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4",
  "access": "open",
  "provider": "hf-inference-providers",
  "task": "pelican_bicycle",
  "sample": 7,
  "reward": 0.8465277777777778,
  "structure_score": 1.0,
  "semantic_score": 0.7638888888888888,
  "gate_passed": true,
  "violations": [],
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "A bird with an orange beak is riding a bicycle.",
  "latency": 101.63,
  "svg": "svg/nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__007.svg",
  "png": "png/nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__007.png"
}
nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__008
nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__008

Upstream outputs, not a Pelican Map ranking · reward 0.801

A cartoon stork is riding a bicycle.

Upstream SVG ↗ · Local SVG

Original fields and blind caption
{
  "model": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4",
  "access": "open",
  "provider": "hf-inference-providers",
  "task": "pelican_bicycle",
  "sample": 8,
  "reward": 0.8013888888888889,
  "structure_score": 1.0,
  "semantic_score": 0.6944444444444444,
  "gate_passed": true,
  "violations": [],
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "A cartoon stork is riding a bicycle.",
  "latency": 12.76,
  "svg": "svg/nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__008.svg",
  "png": "png/nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__008.png"
}
nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__009
nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__009

Upstream outputs, not a Pelican Map ranking · reward 0.950

A cartoon pelican is riding a bicycle.

Upstream SVG ↗ · Local SVG

Original fields and blind caption
{
  "model": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4",
  "access": "open",
  "provider": "hf-inference-providers",
  "task": "pelican_bicycle",
  "sample": 9,
  "reward": 0.95,
  "structure_score": 0.8571428571428571,
  "semantic_score": 1.0,
  "gate_passed": true,
  "violations": [],
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "A cartoon pelican is riding a bicycle.",
  "latency": 10.91,
  "svg": "svg/nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__009.svg",
  "png": "png/nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__009.png"
}
nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__010
nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__010

Upstream outputs, not a Pelican Map ranking · reward 0.494

The image shows a bicycle with a pelican perched on its frame.

Upstream SVG ↗ · Local SVG

Original fields and blind caption
{
  "model": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4",
  "access": "open",
  "provider": "hf-inference-providers",
  "task": "pelican_bicycle",
  "sample": 10,
  "reward": 0.4944444444444444,
  "structure_score": 1.0,
  "semantic_score": 0.2222222222222222,
  "gate_passed": true,
  "violations": [],
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "The image shows a bicycle with a pelican perched on its frame.",
  "latency": 16.29,
  "svg": "svg/nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__010.svg",
  "png": "png/nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__010.png"
}
nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__011
nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__011

Upstream outputs, not a Pelican Map ranking · reward 0.883

A bird with an orange beak is riding a bicycle while holding a paper airplane in its beak.

Upstream SVG ↗ · Local SVG

Original fields and blind caption
{
  "model": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4",
  "access": "open",
  "provider": "hf-inference-providers",
  "task": "pelican_bicycle",
  "sample": 11,
  "reward": 0.8826388888888889,
  "structure_score": 1.0,
  "semantic_score": 0.8194444444444444,
  "gate_passed": true,
  "violations": [],
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "A bird with an orange beak is riding a bicycle while holding a paper airplane in its beak.",
  "latency": 22.6,
  "svg": "svg/nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__011.svg",
  "png": "png/nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__011.png"
}
nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__012
nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__012

Upstream outputs, not a Pelican Map ranking · reward 0.838

The image depicts a stylized drawing of a bicycle with a large, rounded front and two wheels, but the rider is not clearly identifiable.

Upstream SVG ↗ · Local SVG

Original fields and blind caption
{
  "model": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4",
  "access": "open",
  "provider": "hf-inference-providers",
  "task": "pelican_bicycle",
  "sample": 12,
  "reward": 0.8375,
  "structure_score": 1.0,
  "semantic_score": 0.75,
  "gate_passed": true,
  "violations": [],
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "The image depicts a stylized drawing of a bicycle with a large, rounded front and two wheels, but the rider is not clearly identifiable.",
  "latency": 10.56,
  "svg": "svg/nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__012.svg",
  "png": "png/nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__012.png"
}
nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__013
nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__013

Upstream outputs, not a Pelican Map ranking · reward 0.914

A cartoon pelican is riding a bicycle with its wings spread out.

Upstream SVG ↗ · Local SVG

Original fields and blind caption
{
  "model": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4",
  "access": "open",
  "provider": "hf-inference-providers",
  "task": "pelican_bicycle",
  "sample": 13,
  "reward": 0.913888888888889,
  "structure_score": 0.8571428571428571,
  "semantic_score": 0.9444444444444444,
  "gate_passed": true,
  "violations": [],
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "A cartoon pelican is riding a bicycle with its wings spread out.",
  "latency": 129.18,
  "svg": "svg/nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__013.svg",
  "png": "png/nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__013.png"
}
nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__014
nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__014

Upstream outputs, not a Pelican Map ranking · reward 1.000

A cartoon pelican is riding a bicycle.

Upstream SVG ↗ · Local SVG

Original fields and blind caption
{
  "model": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4",
  "access": "open",
  "provider": "hf-inference-providers",
  "task": "pelican_bicycle",
  "sample": 14,
  "reward": 1.0,
  "structure_score": 1.0,
  "semantic_score": 1.0,
  "gate_passed": true,
  "violations": [],
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "A cartoon pelican is riding a bicycle.",
  "latency": 21.0,
  "svg": "svg/nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__014.svg",
  "png": "png/nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__014.png"
}
nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__015
nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__015

Upstream outputs, not a Pelican Map ranking · reward 0.456

A bicycle with orange arrows pointing at its pedals and handlebars, and a bird-like figure above it.

Upstream SVG ↗ · Local SVG

Original fields and blind caption
{
  "model": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4",
  "access": "open",
  "provider": "hf-inference-providers",
  "task": "pelican_bicycle",
  "sample": 15,
  "reward": 0.4560763888888889,
  "structure_score": 1.0,
  "semantic_score": 0.16319444444444445,
  "gate_passed": true,
  "violations": [],
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "A bicycle with orange arrows pointing at its pedals and handlebars, and a bird-like figure above it.",
  "latency": 296.77,
  "svg": "svg/nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__015.svg",
  "png": "png/nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__015.png"
}
nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__016
nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__016

Upstream outputs, not a Pelican Map ranking · reward 0.838

A cartoon duck is riding a bicycle.

Upstream SVG ↗ · Local SVG

Original fields and blind caption
{
  "model": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4",
  "access": "open",
  "provider": "hf-inference-providers",
  "task": "pelican_bicycle",
  "sample": 16,
  "reward": 0.8375,
  "structure_score": 1.0,
  "semantic_score": 0.75,
  "gate_passed": true,
  "violations": [],
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "A cartoon duck is riding a bicycle.",
  "latency": 16.48,
  "svg": "svg/nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__016.svg",
  "png": "png/nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__016.png"
}
nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__017
nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__017

Upstream outputs, not a Pelican Map ranking · reward 0.964

A cartoon pelican is riding a bicycle with large wheels on a grassy area against a light blue background.

Upstream SVG ↗ · Local SVG

Original fields and blind caption
{
  "model": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4",
  "access": "open",
  "provider": "hf-inference-providers",
  "task": "pelican_bicycle",
  "sample": 17,
  "reward": 0.9638888888888889,
  "structure_score": 1.0,
  "semantic_score": 0.9444444444444444,
  "gate_passed": true,
  "violations": [],
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "A cartoon pelican is riding a bicycle with large wheels on a grassy area against a light blue background.",
  "latency": 24.27,
  "svg": "svg/nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__017.svg",
  "png": "png/nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__017.png"
}
nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__018
nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__018

Upstream outputs, not a Pelican Map ranking · reward 1.000

A cartoon pelican is riding a bicycle on a line.

Upstream SVG ↗ · Local SVG

Original fields and blind caption
{
  "model": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4",
  "access": "open",
  "provider": "hf-inference-providers",
  "task": "pelican_bicycle",
  "sample": 18,
  "reward": 1.0,
  "structure_score": 1.0,
  "semantic_score": 1.0,
  "gate_passed": true,
  "violations": [],
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "A cartoon pelican is riding a bicycle on a line.",
  "latency": 17.09,
  "svg": "svg/nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__018.svg",
  "png": "png/nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__018.png"
}
nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__019
nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__019

Upstream outputs, not a Pelican Map ranking · reward 0.810

A cartoon bird with an orange beak is riding a bicycle.

Upstream SVG ↗ · Local SVG

Original fields and blind caption
{
  "model": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4",
  "access": "open",
  "provider": "hf-inference-providers",
  "task": "pelican_bicycle",
  "sample": 19,
  "reward": 0.8104166666666666,
  "structure_score": 1.0,
  "semantic_score": 0.7083333333333333,
  "gate_passed": true,
  "violations": [],
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "A cartoon bird with an orange beak is riding a bicycle.",
  "latency": 10.86,
  "svg": "svg/nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__019.svg",
  "png": "png/nvidia__NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4__pelican_bicycle__019.png"
}

← Back to scoring reference