OpenAI: o4 Mini (batch)
OpenAI o4-mini is a compact reasoning model in the o-series, optimized for fast, cost-efficient performance while retaining strong multimodal and agentic capabilities. It supports tool use and demons…
openai/o4-mini-2025-04-16
OpenAI o4-mini is a compact reasoning model in the o-series, optimized for fast, cost-efficient performance while retaining strong multimodal and agentic capabilities. It supports tool use and demonstrates competitive reasoning and coding performance across benchmarks like AIME (99.5% with Python) and SWE-bench, outperforming its predecessor o3-mini and even approaching o3 in some domains.
Despite its smaller size, o4-mini exhibits high accuracy in STEM tasks, visual problem solving (e.g., MathVista, MMMU), and code editing. It is especially well-suited for high-throughput scenarios where latency or cost is critical. Thanks to its efficient architecture and refined reinforcement learning training, o4-mini can chain tools, generate structured outputs, and solve multi-step tasks with minimal delay—often in under a minute.
Specifications & capabilities
- Input
- image, text, file
- Output
- text
- Context window · tokens
- 200000
- Published model ID
- o4-mini-2025-04-16
- Capabilities
- fileInput, multimodal, promptCaching, reasoning, structuredOutputs, toolCalling, visionInput, webSearch
Description source ↗Evaluations
21–40 / 912
100%
- source id
- kaggleBenchmarks
- leaderboard name
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-093
- dimension
- reasoning
- metric
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-093
- rank
- 1
- aggregation scope
- kaggle
- true
- status
- published
- license
- Apache-2.0
- direction
- higherIsBetter
- conditions
- task
- id
- 1071496
- name
- gpqa-main-zero-shot/task-093
- version
- 1
- sort_order
- DESCENDING
- display_type
- percentages
- benchmark
- id
- 8
- name
- GPQA Main Zero‑Shot
- type
- INDIVIDUAL
- owner
- open-benchmarks
- version
- 2
- categories
- general knowledge and reasoning
- science
- version_id
- 8463
- maintainer_tier
- Kaggle-maintained
- maintenance_level
- KAGGLE_MAINTAINED
- candidate
- candidate_class
- modelVersion
- record_kind
- aggregate
- attribution
- Kaggle · open-benchmarks · GPQA Main Zero‑Shot
- license_url
- https://www.apache.org/licenses/LICENSE-2.0
- metric_kind
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-093
- rank_method
- competition rank among reported evaluations in identical conditions, including unlinked observations
- source_file
- 8463/leaderboard.json
- source_page
- https://www.kaggle.com/benchmarks/open-benchmarks/gpqa-zero-shot-main-set/versions/2
- measurements
- name
- gpqa-main-zero-shot/task-093
- unit
- percent
- scale
- 1
- metric
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-093
- result
- resultCase
- numericResult
- booleanResult
- false
- numericResult
- value
- 1
- standardDeviation
- 0
- confidenceInterval
- 0
- hasStandardDeviation
- false
- hasConfidenceInterval
- false
- hasUnevenConfidenceInterval
- false
- taskVersionId
- 0
- evaluationDate
- 2026-10-20T18:58:28.4490128Z
- hasBooleanResult
- false
- hasNumericResult
- true
- hasTaskVersionId
- false
- hasEvaluationDate
- true
- numericResultNullable
- value
- 1
- standardDeviation
- 0
- confidenceInterval
- 0
- hasStandardDeviation
- false
- hasConfidenceInterval
- false
- hasUnevenConfidenceInterval
- false
- hasNumericResultPublic
- false
- hasNumericResultPrivate
- false
- task_id
- 1071496
- sort_order
- DESCENDING
- source_name
- gpqa-main-zero-shot/task-093
- display_type
- percentages
- task_version
- 1
- task_display_type
- RAW_VALUES
- normalization_status
- normalized
- published_at
- 2026-10-21T18:15:21.342276500Z
- candidate_class
- modelVersion
- source_revision
- b32d007df5acbddafca23c7dd2beba8221c519cf7dd3445a1d7bc28740ffd0d8
- publisher_anchor
- openai/o4-mini-2025-04-16
- metric_population
- 26
- benchmark_evaluation
- true
- task_metadata_source
- benchmarkVersion
- candidate_metadata_source
- benchmarkVersionMappings
- evaluated at
- 2026-10-20T18:58:28.449Z
- source model id
- o4 mini
Original result ↗100%
- source id
- kaggleBenchmarks
- leaderboard name
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-327
- dimension
- reasoning
- metric
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-327
- rank
- 1
- aggregation scope
- kaggle
- true
- status
- published
- license
- Apache-2.0
- direction
- higherIsBetter
- conditions
- task
- id
- 1071754
- name
- gpqa-main-zero-shot/task-327
- version
- 1
- sort_order
- DESCENDING
- display_type
- percentages
- benchmark
- id
- 8
- name
- GPQA Main Zero‑Shot
- type
- INDIVIDUAL
- owner
- open-benchmarks
- version
- 2
- categories
- general knowledge and reasoning
- science
- version_id
- 8463
- maintainer_tier
- Kaggle-maintained
- maintenance_level
- KAGGLE_MAINTAINED
- candidate
- candidate_class
- modelVersion
- record_kind
- aggregate
- attribution
- Kaggle · open-benchmarks · GPQA Main Zero‑Shot
- license_url
- https://www.apache.org/licenses/LICENSE-2.0
- metric_kind
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-327
- rank_method
- competition rank among reported evaluations in identical conditions, including unlinked observations
- source_file
- 8463/leaderboard.json
- source_page
- https://www.kaggle.com/benchmarks/open-benchmarks/gpqa-zero-shot-main-set/versions/2
- measurements
- name
- gpqa-main-zero-shot/task-327
- unit
- percent
- scale
- 1
- metric
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-327
- result
- resultCase
- numericResult
- booleanResult
- false
- numericResult
- value
- 1
- standardDeviation
- 0
- confidenceInterval
- 0
- hasStandardDeviation
- false
- hasConfidenceInterval
- false
- hasUnevenConfidenceInterval
- false
- taskVersionId
- 0
- evaluationDate
- 2026-10-20T20:46:03.039921Z
- hasBooleanResult
- false
- hasNumericResult
- true
- hasTaskVersionId
- false
- hasEvaluationDate
- true
- numericResultNullable
- value
- 1
- standardDeviation
- 0
- confidenceInterval
- 0
- hasStandardDeviation
- false
- hasConfidenceInterval
- false
- hasUnevenConfidenceInterval
- false
- hasNumericResultPublic
- false
- hasNumericResultPrivate
- false
- task_id
- 1071754
- sort_order
- DESCENDING
- source_name
- gpqa-main-zero-shot/task-327
- display_type
- percentages
- task_version
- 1
- task_display_type
- RAW_VALUES
- normalization_status
- normalized
- published_at
- 2026-10-21T18:15:21.342276500Z
- candidate_class
- modelVersion
- source_revision
- b32d007df5acbddafca23c7dd2beba8221c519cf7dd3445a1d7bc28740ffd0d8
- publisher_anchor
- openai/o4-mini-2025-04-16
- metric_population
- 26
- benchmark_evaluation
- true
- task_metadata_source
- benchmarkVersion
- candidate_metadata_source
- benchmarkVersionMappings
- evaluated at
- 2026-10-20T20:46:03.039Z
- source model id
- o4 mini
Original result ↗100%
- source id
- kaggleBenchmarks
- leaderboard name
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-168
- dimension
- reasoning
- metric
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-168
- rank
- 1
- aggregation scope
- kaggle
- true
- status
- published
- license
- Apache-2.0
- direction
- higherIsBetter
- conditions
- task
- id
- 1071581
- name
- gpqa-main-zero-shot/task-168
- version
- 1
- sort_order
- DESCENDING
- display_type
- percentages
- benchmark
- id
- 8
- name
- GPQA Main Zero‑Shot
- type
- INDIVIDUAL
- owner
- open-benchmarks
- version
- 2
- categories
- general knowledge and reasoning
- science
- version_id
- 8463
- maintainer_tier
- Kaggle-maintained
- maintenance_level
- KAGGLE_MAINTAINED
- candidate
- candidate_class
- modelVersion
- record_kind
- aggregate
- attribution
- Kaggle · open-benchmarks · GPQA Main Zero‑Shot
- license_url
- https://www.apache.org/licenses/LICENSE-2.0
- metric_kind
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-168
- rank_method
- competition rank among reported evaluations in identical conditions, including unlinked observations
- source_file
- 8463/leaderboard.json
- source_page
- https://www.kaggle.com/benchmarks/open-benchmarks/gpqa-zero-shot-main-set/versions/2
- measurements
- name
- gpqa-main-zero-shot/task-168
- unit
- percent
- scale
- 1
- metric
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-168
- result
- resultCase
- numericResult
- booleanResult
- false
- numericResult
- value
- 1
- standardDeviation
- 0
- confidenceInterval
- 0
- hasStandardDeviation
- false
- hasConfidenceInterval
- false
- hasUnevenConfidenceInterval
- false
- taskVersionId
- 0
- evaluationDate
- 2026-10-20T19:34:19.7769117Z
- hasBooleanResult
- false
- hasNumericResult
- true
- hasTaskVersionId
- false
- hasEvaluationDate
- true
- numericResultNullable
- value
- 1
- standardDeviation
- 0
- confidenceInterval
- 0
- hasStandardDeviation
- false
- hasConfidenceInterval
- false
- hasUnevenConfidenceInterval
- false
- hasNumericResultPublic
- false
- hasNumericResultPrivate
- false
- task_id
- 1071581
- sort_order
- DESCENDING
- source_name
- gpqa-main-zero-shot/task-168
- display_type
- percentages
- task_version
- 1
- task_display_type
- RAW_VALUES
- normalization_status
- normalized
- published_at
- 2026-10-21T18:15:21.342276500Z
- candidate_class
- modelVersion
- source_revision
- b32d007df5acbddafca23c7dd2beba8221c519cf7dd3445a1d7bc28740ffd0d8
- publisher_anchor
- openai/o4-mini-2025-04-16
- metric_population
- 26
- benchmark_evaluation
- true
- task_metadata_source
- benchmarkVersion
- candidate_metadata_source
- benchmarkVersionMappings
- evaluated at
- 2026-10-20T19:34:19.776Z
- source model id
- o4 mini
Original result ↗100%
- source id
- kaggleBenchmarks
- leaderboard name
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-103
- dimension
- reasoning
- metric
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-103
- rank
- 1
- aggregation scope
- kaggle
- true
- status
- published
- license
- Apache-2.0
- direction
- higherIsBetter
- conditions
- task
- id
- 1071506
- name
- gpqa-main-zero-shot/task-103
- version
- 1
- sort_order
- DESCENDING
- display_type
- percentages
- benchmark
- id
- 8
- name
- GPQA Main Zero‑Shot
- type
- INDIVIDUAL
- owner
- open-benchmarks
- version
- 2
- categories
- general knowledge and reasoning
- science
- version_id
- 8463
- maintainer_tier
- Kaggle-maintained
- maintenance_level
- KAGGLE_MAINTAINED
- candidate
- candidate_class
- modelVersion
- record_kind
- aggregate
- attribution
- Kaggle · open-benchmarks · GPQA Main Zero‑Shot
- license_url
- https://www.apache.org/licenses/LICENSE-2.0
- metric_kind
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-103
- rank_method
- competition rank among reported evaluations in identical conditions, including unlinked observations
- source_file
- 8463/leaderboard.json
- source_page
- https://www.kaggle.com/benchmarks/open-benchmarks/gpqa-zero-shot-main-set/versions/2
- measurements
- name
- gpqa-main-zero-shot/task-103
- unit
- percent
- scale
- 1
- metric
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-103
- result
- resultCase
- numericResult
- booleanResult
- false
- numericResult
- value
- 1
- standardDeviation
- 0
- confidenceInterval
- 0
- hasStandardDeviation
- false
- hasConfidenceInterval
- false
- hasUnevenConfidenceInterval
- false
- taskVersionId
- 0
- evaluationDate
- 2026-10-20T19:02:21.8744508Z
- hasBooleanResult
- false
- hasNumericResult
- true
- hasTaskVersionId
- false
- hasEvaluationDate
- true
- numericResultNullable
- value
- 1
- standardDeviation
- 0
- confidenceInterval
- 0
- hasStandardDeviation
- false
- hasConfidenceInterval
- false
- hasUnevenConfidenceInterval
- false
- hasNumericResultPublic
- false
- hasNumericResultPrivate
- false
- task_id
- 1071506
- sort_order
- DESCENDING
- source_name
- gpqa-main-zero-shot/task-103
- display_type
- percentages
- task_version
- 1
- task_display_type
- RAW_VALUES
- normalization_status
- normalized
- published_at
- 2026-10-21T18:15:21.342276500Z
- candidate_class
- modelVersion
- source_revision
- b32d007df5acbddafca23c7dd2beba8221c519cf7dd3445a1d7bc28740ffd0d8
- publisher_anchor
- openai/o4-mini-2025-04-16
- metric_population
- 26
- benchmark_evaluation
- true
- task_metadata_source
- benchmarkVersion
- candidate_metadata_source
- benchmarkVersionMappings
- evaluated at
- 2026-10-20T19:02:21.874Z
- source model id
- o4 mini
Original result ↗100%
- source id
- kaggleBenchmarks
- leaderboard name
- GPQA Diamond Zero‑Shot · gpqa-diamond-zero-shot/task-054
- dimension
- reasoning
- metric
- GPQA Diamond Zero‑Shot · gpqa-diamond-zero-shot/task-054
- rank
- 1
- aggregation scope
- kaggle
- true
- status
- published
- license
- Apache-2.0
- direction
- higherIsBetter
- conditions
- task
- id
- 984103
- name
- gpqa-diamond-zero-shot/task-054
- version
- 1
- sort_order
- DESCENDING
- display_type
- percentages
- benchmark
- id
- 7
- name
- GPQA Diamond Zero‑Shot
- type
- INDIVIDUAL
- owner
- open-benchmarks
- version
- 2
- categories
- general knowledge and reasoning
- science
- version_id
- 8176
- maintainer_tier
- Kaggle-maintained
- maintenance_level
- KAGGLE_MAINTAINED
- candidate
- candidate_class
- modelVersion
- record_kind
- aggregate
- attribution
- Kaggle · open-benchmarks · GPQA Diamond Zero‑Shot
- license_url
- https://www.apache.org/licenses/LICENSE-2.0
- metric_kind
- GPQA Diamond Zero‑Shot · gpqa-diamond-zero-shot/task-054
- rank_method
- competition rank among reported evaluations in identical conditions, including unlinked observations
- source_file
- 8176/leaderboard.json
- source_page
- https://www.kaggle.com/benchmarks/open-benchmarks/gpqa-zero-shot-diamond-set/versions/2
- measurements
- name
- gpqa-diamond-zero-shot/task-054
- unit
- percent
- scale
- 1
- metric
- GPQA Diamond Zero‑Shot · gpqa-diamond-zero-shot/task-054
- result
- resultCase
- numericResult
- booleanResult
- false
- numericResult
- value
- 1
- standardDeviation
- 0
- confidenceInterval
- 0
- hasStandardDeviation
- false
- hasConfidenceInterval
- false
- hasUnevenConfidenceInterval
- false
- taskVersionId
- 0
- evaluationDate
- 2026-10-12T18:11:45.7010661Z
- hasBooleanResult
- false
- hasNumericResult
- true
- hasTaskVersionId
- false
- hasEvaluationDate
- true
- numericResultNullable
- value
- 1
- standardDeviation
- 0
- confidenceInterval
- 0
- hasStandardDeviation
- false
- hasConfidenceInterval
- false
- hasUnevenConfidenceInterval
- false
- hasNumericResultPublic
- false
- hasNumericResultPrivate
- false
- task_id
- 984103
- sort_order
- DESCENDING
- source_name
- gpqa-diamond-zero-shot/task-054
- display_type
- percentages
- task_version
- 1
- task_display_type
- RAW_VALUES
- normalization_status
- normalized
- published_at
- 2026-10-12T17:49:58.132967500Z
- candidate_class
- modelVersion
- source_revision
- b32d007df5acbddafca23c7dd2beba8221c519cf7dd3445a1d7bc28740ffd0d8
- publisher_anchor
- openai/o4-mini-2025-04-16
- metric_population
- 32
- benchmark_evaluation
- true
- task_metadata_source
- benchmarkVersion
- candidate_metadata_source
- benchmarkVersionMappings
- evaluated at
- 2026-10-12T18:11:45.701Z
- source model id
- o4 mini
Original result ↗100%
- source id
- kaggleBenchmarks
- leaderboard name
- GPQA Diamond Zero‑Shot · gpqa-diamond-zero-shot/task-117
- dimension
- reasoning
- metric
- GPQA Diamond Zero‑Shot · gpqa-diamond-zero-shot/task-117
- rank
- 1
- aggregation scope
- kaggle
- true
- status
- published
- license
- Apache-2.0
- direction
- higherIsBetter
- conditions
- task
- id
- 984166
- name
- gpqa-diamond-zero-shot/task-117
- version
- 1
- sort_order
- DESCENDING
- display_type
- percentages
- benchmark
- id
- 7
- name
- GPQA Diamond Zero‑Shot
- type
- INDIVIDUAL
- owner
- open-benchmarks
- version
- 2
- categories
- general knowledge and reasoning
- science
- version_id
- 8176
- maintainer_tier
- Kaggle-maintained
- maintenance_level
- KAGGLE_MAINTAINED
- candidate
- candidate_class
- modelVersion
- record_kind
- aggregate
- attribution
- Kaggle · open-benchmarks · GPQA Diamond Zero‑Shot
- license_url
- https://www.apache.org/licenses/LICENSE-2.0
- metric_kind
- GPQA Diamond Zero‑Shot · gpqa-diamond-zero-shot/task-117
- rank_method
- competition rank among reported evaluations in identical conditions, including unlinked observations
- source_file
- 8176/leaderboard.json
- source_page
- https://www.kaggle.com/benchmarks/open-benchmarks/gpqa-zero-shot-diamond-set/versions/2
- measurements
- name
- gpqa-diamond-zero-shot/task-117
- unit
- percent
- scale
- 1
- metric
- GPQA Diamond Zero‑Shot · gpqa-diamond-zero-shot/task-117
- result
- resultCase
- numericResult
- booleanResult
- false
- numericResult
- value
- 1
- standardDeviation
- 0
- confidenceInterval
- 0
- hasStandardDeviation
- false
- hasConfidenceInterval
- false
- hasUnevenConfidenceInterval
- false
- taskVersionId
- 0
- evaluationDate
- 2026-10-12T18:13:34.6756324Z
- hasBooleanResult
- false
- hasNumericResult
- true
- hasTaskVersionId
- false
- hasEvaluationDate
- true
- numericResultNullable
- value
- 1
- standardDeviation
- 0
- confidenceInterval
- 0
- hasStandardDeviation
- false
- hasConfidenceInterval
- false
- hasUnevenConfidenceInterval
- false
- hasNumericResultPublic
- false
- hasNumericResultPrivate
- false
- task_id
- 984166
- sort_order
- DESCENDING
- source_name
- gpqa-diamond-zero-shot/task-117
- display_type
- percentages
- task_version
- 1
- task_display_type
- RAW_VALUES
- normalization_status
- normalized
- published_at
- 2026-10-12T17:49:58.132967500Z
- candidate_class
- modelVersion
- source_revision
- b32d007df5acbddafca23c7dd2beba8221c519cf7dd3445a1d7bc28740ffd0d8
- publisher_anchor
- openai/o4-mini-2025-04-16
- metric_population
- 32
- benchmark_evaluation
- true
- task_metadata_source
- benchmarkVersion
- candidate_metadata_source
- benchmarkVersionMappings
- evaluated at
- 2026-10-12T18:13:34.675Z
- source model id
- o4 mini
Original result ↗100%
- source id
- kaggleBenchmarks
- leaderboard name
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-321
- dimension
- reasoning
- metric
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-321
- rank
- 1
- aggregation scope
- kaggle
- true
- status
- published
- license
- Apache-2.0
- direction
- higherIsBetter
- conditions
- task
- id
- 1071748
- name
- gpqa-main-zero-shot/task-321
- version
- 1
- sort_order
- DESCENDING
- display_type
- percentages
- benchmark
- id
- 8
- name
- GPQA Main Zero‑Shot
- type
- INDIVIDUAL
- owner
- open-benchmarks
- version
- 2
- categories
- general knowledge and reasoning
- science
- version_id
- 8463
- maintainer_tier
- Kaggle-maintained
- maintenance_level
- KAGGLE_MAINTAINED
- candidate
- candidate_class
- modelVersion
- record_kind
- aggregate
- attribution
- Kaggle · open-benchmarks · GPQA Main Zero‑Shot
- license_url
- https://www.apache.org/licenses/LICENSE-2.0
- metric_kind
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-321
- rank_method
- competition rank among reported evaluations in identical conditions, including unlinked observations
- source_file
- 8463/leaderboard.json
- source_page
- https://www.kaggle.com/benchmarks/open-benchmarks/gpqa-zero-shot-main-set/versions/2
- measurements
- name
- gpqa-main-zero-shot/task-321
- unit
- percent
- scale
- 1
- metric
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-321
- result
- resultCase
- numericResult
- booleanResult
- false
- numericResult
- value
- 1
- standardDeviation
- 0
- confidenceInterval
- 0
- hasStandardDeviation
- false
- hasConfidenceInterval
- false
- hasUnevenConfidenceInterval
- false
- taskVersionId
- 0
- evaluationDate
- 2026-10-20T20:43:50.5058058Z
- hasBooleanResult
- false
- hasNumericResult
- true
- hasTaskVersionId
- false
- hasEvaluationDate
- true
- numericResultNullable
- value
- 1
- standardDeviation
- 0
- confidenceInterval
- 0
- hasStandardDeviation
- false
- hasConfidenceInterval
- false
- hasUnevenConfidenceInterval
- false
- hasNumericResultPublic
- false
- hasNumericResultPrivate
- false
- task_id
- 1071748
- sort_order
- DESCENDING
- source_name
- gpqa-main-zero-shot/task-321
- display_type
- percentages
- task_version
- 1
- task_display_type
- RAW_VALUES
- normalization_status
- normalized
- published_at
- 2026-10-21T18:15:21.342276500Z
- candidate_class
- modelVersion
- source_revision
- b32d007df5acbddafca23c7dd2beba8221c519cf7dd3445a1d7bc28740ffd0d8
- publisher_anchor
- openai/o4-mini-2025-04-16
- metric_population
- 26
- benchmark_evaluation
- true
- task_metadata_source
- benchmarkVersion
- candidate_metadata_source
- benchmarkVersionMappings
- evaluated at
- 2026-10-20T20:43:50.505Z
- source model id
- o4 mini
Original result ↗100%
- source id
- kaggleBenchmarks
- leaderboard name
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-324
- dimension
- reasoning
- metric
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-324
- rank
- 1
- aggregation scope
- kaggle
- true
- status
- published
- license
- Apache-2.0
- direction
- higherIsBetter
- conditions
- task
- id
- 1071751
- name
- gpqa-main-zero-shot/task-324
- version
- 1
- sort_order
- DESCENDING
- display_type
- percentages
- benchmark
- id
- 8
- name
- GPQA Main Zero‑Shot
- type
- INDIVIDUAL
- owner
- open-benchmarks
- version
- 2
- categories
- general knowledge and reasoning
- science
- version_id
- 8463
- maintainer_tier
- Kaggle-maintained
- maintenance_level
- KAGGLE_MAINTAINED
- candidate
- candidate_class
- modelVersion
- record_kind
- aggregate
- attribution
- Kaggle · open-benchmarks · GPQA Main Zero‑Shot
- license_url
- https://www.apache.org/licenses/LICENSE-2.0
- metric_kind
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-324
- rank_method
- competition rank among reported evaluations in identical conditions, including unlinked observations
- source_file
- 8463/leaderboard.json
- source_page
- https://www.kaggle.com/benchmarks/open-benchmarks/gpqa-zero-shot-main-set/versions/2
- measurements
- name
- gpqa-main-zero-shot/task-324
- unit
- percent
- scale
- 1
- metric
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-324
- result
- resultCase
- numericResult
- booleanResult
- false
- numericResult
- value
- 1
- standardDeviation
- 0
- confidenceInterval
- 0
- hasStandardDeviation
- false
- hasConfidenceInterval
- false
- hasUnevenConfidenceInterval
- false
- taskVersionId
- 0
- evaluationDate
- 2026-10-20T20:45:45.7979807Z
- hasBooleanResult
- false
- hasNumericResult
- true
- hasTaskVersionId
- false
- hasEvaluationDate
- true
- numericResultNullable
- value
- 1
- standardDeviation
- 0
- confidenceInterval
- 0
- hasStandardDeviation
- false
- hasConfidenceInterval
- false
- hasUnevenConfidenceInterval
- false
- hasNumericResultPublic
- false
- hasNumericResultPrivate
- false
- task_id
- 1071751
- sort_order
- DESCENDING
- source_name
- gpqa-main-zero-shot/task-324
- display_type
- percentages
- task_version
- 1
- task_display_type
- RAW_VALUES
- normalization_status
- normalized
- published_at
- 2026-10-21T18:15:21.342276500Z
- candidate_class
- modelVersion
- source_revision
- b32d007df5acbddafca23c7dd2beba8221c519cf7dd3445a1d7bc28740ffd0d8
- publisher_anchor
- openai/o4-mini-2025-04-16
- metric_population
- 26
- benchmark_evaluation
- true
- task_metadata_source
- benchmarkVersion
- candidate_metadata_source
- benchmarkVersionMappings
- evaluated at
- 2026-10-20T20:45:45.797Z
- source model id
- o4 mini
Original result ↗100%
- source id
- kaggleBenchmarks
- leaderboard name
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-184
- dimension
- reasoning
- metric
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-184
- rank
- 1
- aggregation scope
- kaggle
- true
- status
- published
- license
- Apache-2.0
- direction
- higherIsBetter
- conditions
- task
- id
- 1071598
- name
- gpqa-main-zero-shot/task-184
- version
- 1
- sort_order
- DESCENDING
- display_type
- percentages
- benchmark
- id
- 8
- name
- GPQA Main Zero‑Shot
- type
- INDIVIDUAL
- owner
- open-benchmarks
- version
- 2
- categories
- general knowledge and reasoning
- science
- version_id
- 8463
- maintainer_tier
- Kaggle-maintained
- maintenance_level
- KAGGLE_MAINTAINED
- candidate
- candidate_class
- modelVersion
- record_kind
- aggregate
- attribution
- Kaggle · open-benchmarks · GPQA Main Zero‑Shot
- license_url
- https://www.apache.org/licenses/LICENSE-2.0
- metric_kind
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-184
- rank_method
- competition rank among reported evaluations in identical conditions, including unlinked observations
- source_file
- 8463/leaderboard.json
- source_page
- https://www.kaggle.com/benchmarks/open-benchmarks/gpqa-zero-shot-main-set/versions/2
- measurements
- name
- gpqa-main-zero-shot/task-184
- unit
- percent
- scale
- 1
- metric
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-184
- result
- resultCase
- numericResult
- booleanResult
- false
- numericResult
- value
- 1
- standardDeviation
- 0
- confidenceInterval
- 0
- hasStandardDeviation
- false
- hasConfidenceInterval
- false
- hasUnevenConfidenceInterval
- false
- taskVersionId
- 0
- evaluationDate
- 2026-10-20T19:42:15.4422889Z
- hasBooleanResult
- false
- hasNumericResult
- true
- hasTaskVersionId
- false
- hasEvaluationDate
- true
- numericResultNullable
- value
- 1
- standardDeviation
- 0
- confidenceInterval
- 0
- hasStandardDeviation
- false
- hasConfidenceInterval
- false
- hasUnevenConfidenceInterval
- false
- hasNumericResultPublic
- false
- hasNumericResultPrivate
- false
- task_id
- 1071598
- sort_order
- DESCENDING
- source_name
- gpqa-main-zero-shot/task-184
- display_type
- percentages
- task_version
- 1
- task_display_type
- RAW_VALUES
- normalization_status
- normalized
- published_at
- 2026-10-21T18:15:21.342276500Z
- candidate_class
- modelVersion
- source_revision
- b32d007df5acbddafca23c7dd2beba8221c519cf7dd3445a1d7bc28740ffd0d8
- publisher_anchor
- openai/o4-mini-2025-04-16
- metric_population
- 26
- benchmark_evaluation
- true
- task_metadata_source
- benchmarkVersion
- candidate_metadata_source
- benchmarkVersionMappings
- evaluated at
- 2026-10-20T19:42:15.442Z
- source model id
- o4 mini
Original result ↗100%
- source id
- kaggleBenchmarks
- leaderboard name
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-358
- dimension
- reasoning
- metric
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-358
- rank
- 1
- aggregation scope
- kaggle
- true
- status
- published
- license
- Apache-2.0
- direction
- higherIsBetter
- conditions
- task
- id
- 1071789
- name
- gpqa-main-zero-shot/task-358
- version
- 1
- sort_order
- DESCENDING
- display_type
- percentages
- benchmark
- id
- 8
- name
- GPQA Main Zero‑Shot
- type
- INDIVIDUAL
- owner
- open-benchmarks
- version
- 2
- categories
- general knowledge and reasoning
- science
- version_id
- 8463
- maintainer_tier
- Kaggle-maintained
- maintenance_level
- KAGGLE_MAINTAINED
- candidate
- candidate_class
- modelVersion
- record_kind
- aggregate
- attribution
- Kaggle · open-benchmarks · GPQA Main Zero‑Shot
- license_url
- https://www.apache.org/licenses/LICENSE-2.0
- metric_kind
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-358
- rank_method
- competition rank among reported evaluations in identical conditions, including unlinked observations
- source_file
- 8463/leaderboard.json
- source_page
- https://www.kaggle.com/benchmarks/open-benchmarks/gpqa-zero-shot-main-set/versions/2
- measurements
- name
- gpqa-main-zero-shot/task-358
- unit
- percent
- scale
- 1
- metric
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-358
- result
- resultCase
- numericResult
- booleanResult
- false
- numericResult
- value
- 1
- standardDeviation
- 0
- confidenceInterval
- 0
- hasStandardDeviation
- false
- hasConfidenceInterval
- false
- hasUnevenConfidenceInterval
- false
- taskVersionId
- 0
- evaluationDate
- 2026-10-20T20:59:53.0918646Z
- hasBooleanResult
- false
- hasNumericResult
- true
- hasTaskVersionId
- false
- hasEvaluationDate
- true
- numericResultNullable
- value
- 1
- standardDeviation
- 0
- confidenceInterval
- 0
- hasStandardDeviation
- false
- hasConfidenceInterval
- false
- hasUnevenConfidenceInterval
- false
- hasNumericResultPublic
- false
- hasNumericResultPrivate
- false
- task_id
- 1071789
- sort_order
- DESCENDING
- source_name
- gpqa-main-zero-shot/task-358
- display_type
- percentages
- task_version
- 1
- task_display_type
- RAW_VALUES
- normalization_status
- normalized
- published_at
- 2026-10-21T18:15:21.342276500Z
- candidate_class
- modelVersion
- source_revision
- b32d007df5acbddafca23c7dd2beba8221c519cf7dd3445a1d7bc28740ffd0d8
- publisher_anchor
- openai/o4-mini-2025-04-16
- metric_population
- 26
- benchmark_evaluation
- true
- task_metadata_source
- benchmarkVersion
- candidate_metadata_source
- benchmarkVersionMappings
- evaluated at
- 2026-10-20T20:59:53.091Z
- source model id
- o4 mini
Original result ↗100%
- source id
- kaggleBenchmarks
- leaderboard name
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-320
- dimension
- reasoning
- metric
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-320
- rank
- 1
- aggregation scope
- kaggle
- true
- status
- published
- license
- Apache-2.0
- direction
- higherIsBetter
- conditions
- task
- id
- 1071747
- name
- gpqa-main-zero-shot/task-320
- version
- 1
- sort_order
- DESCENDING
- display_type
- percentages
- benchmark
- id
- 8
- name
- GPQA Main Zero‑Shot
- type
- INDIVIDUAL
- owner
- open-benchmarks
- version
- 2
- categories
- general knowledge and reasoning
- science
- version_id
- 8463
- maintainer_tier
- Kaggle-maintained
- maintenance_level
- KAGGLE_MAINTAINED
- candidate
- candidate_class
- modelVersion
- record_kind
- aggregate
- attribution
- Kaggle · open-benchmarks · GPQA Main Zero‑Shot
- license_url
- https://www.apache.org/licenses/LICENSE-2.0
- metric_kind
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-320
- rank_method
- competition rank among reported evaluations in identical conditions, including unlinked observations
- source_file
- 8463/leaderboard.json
- source_page
- https://www.kaggle.com/benchmarks/open-benchmarks/gpqa-zero-shot-main-set/versions/2
- measurements
- name
- gpqa-main-zero-shot/task-320
- unit
- percent
- scale
- 1
- metric
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-320
- result
- resultCase
- numericResult
- booleanResult
- false
- numericResult
- value
- 1
- standardDeviation
- 0
- confidenceInterval
- 0
- hasStandardDeviation
- false
- hasConfidenceInterval
- false
- hasUnevenConfidenceInterval
- false
- taskVersionId
- 0
- evaluationDate
- 2026-10-20T20:43:58.9519382Z
- hasBooleanResult
- false
- hasNumericResult
- true
- hasTaskVersionId
- false
- hasEvaluationDate
- true
- numericResultNullable
- value
- 1
- standardDeviation
- 0
- confidenceInterval
- 0
- hasStandardDeviation
- false
- hasConfidenceInterval
- false
- hasUnevenConfidenceInterval
- false
- hasNumericResultPublic
- false
- hasNumericResultPrivate
- false
- task_id
- 1071747
- sort_order
- DESCENDING
- source_name
- gpqa-main-zero-shot/task-320
- display_type
- percentages
- task_version
- 1
- task_display_type
- RAW_VALUES
- normalization_status
- normalized
- published_at
- 2026-10-21T18:15:21.342276500Z
- candidate_class
- modelVersion
- source_revision
- b32d007df5acbddafca23c7dd2beba8221c519cf7dd3445a1d7bc28740ffd0d8
- publisher_anchor
- openai/o4-mini-2025-04-16
- metric_population
- 26
- benchmark_evaluation
- true
- task_metadata_source
- benchmarkVersion
- candidate_metadata_source
- benchmarkVersionMappings
- evaluated at
- 2026-10-20T20:43:58.951Z
- source model id
- o4 mini
Original result ↗100%
- source id
- kaggleBenchmarks
- leaderboard name
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-252
- dimension
- reasoning
- metric
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-252
- rank
- 1
- aggregation scope
- kaggle
- true
- status
- published
- license
- Apache-2.0
- direction
- higherIsBetter
- conditions
- task
- id
- 1071674
- name
- gpqa-main-zero-shot/task-252
- version
- 1
- sort_order
- DESCENDING
- display_type
- percentages
- benchmark
- id
- 8
- name
- GPQA Main Zero‑Shot
- type
- INDIVIDUAL
- owner
- open-benchmarks
- version
- 2
- categories
- general knowledge and reasoning
- science
- version_id
- 8463
- maintainer_tier
- Kaggle-maintained
- maintenance_level
- KAGGLE_MAINTAINED
- candidate
- candidate_class
- modelVersion
- record_kind
- aggregate
- attribution
- Kaggle · open-benchmarks · GPQA Main Zero‑Shot
- license_url
- https://www.apache.org/licenses/LICENSE-2.0
- metric_kind
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-252
- rank_method
- competition rank among reported evaluations in identical conditions, including unlinked observations
- source_file
- 8463/leaderboard.json
- source_page
- https://www.kaggle.com/benchmarks/open-benchmarks/gpqa-zero-shot-main-set/versions/2
- measurements
- name
- gpqa-main-zero-shot/task-252
- unit
- percent
- scale
- 1
- metric
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-252
- result
- resultCase
- numericResult
- booleanResult
- false
- numericResult
- value
- 1
- standardDeviation
- 0
- confidenceInterval
- 0
- hasStandardDeviation
- false
- hasConfidenceInterval
- false
- hasUnevenConfidenceInterval
- false
- taskVersionId
- 0
- evaluationDate
- 2026-10-20T20:12:08.2834609Z
- hasBooleanResult
- false
- hasNumericResult
- true
- hasTaskVersionId
- false
- hasEvaluationDate
- true
- numericResultNullable
- value
- 1
- standardDeviation
- 0
- confidenceInterval
- 0
- hasStandardDeviation
- false
- hasConfidenceInterval
- false
- hasUnevenConfidenceInterval
- false
- hasNumericResultPublic
- false
- hasNumericResultPrivate
- false
- task_id
- 1071674
- sort_order
- DESCENDING
- source_name
- gpqa-main-zero-shot/task-252
- display_type
- percentages
- task_version
- 1
- task_display_type
- RAW_VALUES
- normalization_status
- normalized
- published_at
- 2026-10-21T18:15:21.342276500Z
- candidate_class
- modelVersion
- source_revision
- b32d007df5acbddafca23c7dd2beba8221c519cf7dd3445a1d7bc28740ffd0d8
- publisher_anchor
- openai/o4-mini-2025-04-16
- metric_population
- 26
- benchmark_evaluation
- true
- task_metadata_source
- benchmarkVersion
- candidate_metadata_source
- benchmarkVersionMappings
- evaluated at
- 2026-10-20T20:12:08.283Z
- source model id
- o4 mini
Original result ↗100%
- source id
- kaggleBenchmarks
- leaderboard name
- GPQA Diamond Zero‑Shot · gpqa-diamond-zero-shot/task-163
- dimension
- reasoning
- metric
- GPQA Diamond Zero‑Shot · gpqa-diamond-zero-shot/task-163
- rank
- 1
- aggregation scope
- kaggle
- true
- status
- published
- license
- Apache-2.0
- direction
- higherIsBetter
- conditions
- task
- id
- 984213
- name
- gpqa-diamond-zero-shot/task-163
- version
- 1
- sort_order
- DESCENDING
- display_type
- percentages
- benchmark
- id
- 7
- name
- GPQA Diamond Zero‑Shot
- type
- INDIVIDUAL
- owner
- open-benchmarks
- version
- 2
- categories
- general knowledge and reasoning
- science
- version_id
- 8176
- maintainer_tier
- Kaggle-maintained
- maintenance_level
- KAGGLE_MAINTAINED
- candidate
- candidate_class
- modelVersion
- record_kind
- aggregate
- attribution
- Kaggle · open-benchmarks · GPQA Diamond Zero‑Shot
- license_url
- https://www.apache.org/licenses/LICENSE-2.0
- metric_kind
- GPQA Diamond Zero‑Shot · gpqa-diamond-zero-shot/task-163
- rank_method
- competition rank among reported evaluations in identical conditions, including unlinked observations
- source_file
- 8176/leaderboard.json
- source_page
- https://www.kaggle.com/benchmarks/open-benchmarks/gpqa-zero-shot-diamond-set/versions/2
- measurements
- name
- gpqa-diamond-zero-shot/task-163
- unit
- percent
- scale
- 1
- metric
- GPQA Diamond Zero‑Shot · gpqa-diamond-zero-shot/task-163
- result
- resultCase
- numericResult
- booleanResult
- false
- numericResult
- value
- 1
- standardDeviation
- 0
- confidenceInterval
- 0
- hasStandardDeviation
- false
- hasConfidenceInterval
- false
- hasUnevenConfidenceInterval
- false
- taskVersionId
- 0
- evaluationDate
- 2026-10-12T18:15:30.9926363Z
- hasBooleanResult
- false
- hasNumericResult
- true
- hasTaskVersionId
- false
- hasEvaluationDate
- true
- numericResultNullable
- value
- 1
- standardDeviation
- 0
- confidenceInterval
- 0
- hasStandardDeviation
- false
- hasConfidenceInterval
- false
- hasUnevenConfidenceInterval
- false
- hasNumericResultPublic
- false
- hasNumericResultPrivate
- false
- task_id
- 984213
- sort_order
- DESCENDING
- source_name
- gpqa-diamond-zero-shot/task-163
- display_type
- percentages
- task_version
- 1
- task_display_type
- RAW_VALUES
- normalization_status
- normalized
- published_at
- 2026-10-12T17:49:58.132967500Z
- candidate_class
- modelVersion
- source_revision
- b32d007df5acbddafca23c7dd2beba8221c519cf7dd3445a1d7bc28740ffd0d8
- publisher_anchor
- openai/o4-mini-2025-04-16
- metric_population
- 32
- benchmark_evaluation
- true
- task_metadata_source
- benchmarkVersion
- candidate_metadata_source
- benchmarkVersionMappings
- evaluated at
- 2026-10-12T18:15:30.992Z
- source model id
- o4 mini
Original result ↗100%
- source id
- kaggleBenchmarks
- leaderboard name
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-068
- dimension
- reasoning
- metric
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-068
- rank
- 1
- aggregation scope
- kaggle
- true
- status
- published
- license
- Apache-2.0
- direction
- higherIsBetter
- conditions
- task
- id
- 1071469
- name
- gpqa-main-zero-shot/task-068
- version
- 1
- sort_order
- DESCENDING
- display_type
- percentages
- benchmark
- id
- 8
- name
- GPQA Main Zero‑Shot
- type
- INDIVIDUAL
- owner
- open-benchmarks
- version
- 2
- categories
- general knowledge and reasoning
- science
- version_id
- 8463
- maintainer_tier
- Kaggle-maintained
- maintenance_level
- KAGGLE_MAINTAINED
- candidate
- candidate_class
- modelVersion
- record_kind
- aggregate
- attribution
- Kaggle · open-benchmarks · GPQA Main Zero‑Shot
- license_url
- https://www.apache.org/licenses/LICENSE-2.0
- metric_kind
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-068
- rank_method
- competition rank among reported evaluations in identical conditions, including unlinked observations
- source_file
- 8463/leaderboard.json
- source_page
- https://www.kaggle.com/benchmarks/open-benchmarks/gpqa-zero-shot-main-set/versions/2
- measurements
- name
- gpqa-main-zero-shot/task-068
- unit
- percent
- scale
- 1
- metric
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-068
- result
- resultCase
- numericResult
- booleanResult
- false
- numericResult
- value
- 1
- standardDeviation
- 0
- confidenceInterval
- 0
- hasStandardDeviation
- false
- hasConfidenceInterval
- false
- hasUnevenConfidenceInterval
- false
- taskVersionId
- 0
- evaluationDate
- 2026-10-20T18:39:49.4684688Z
- hasBooleanResult
- false
- hasNumericResult
- true
- hasTaskVersionId
- false
- hasEvaluationDate
- true
- numericResultNullable
- value
- 1
- standardDeviation
- 0
- confidenceInterval
- 0
- hasStandardDeviation
- false
- hasConfidenceInterval
- false
- hasUnevenConfidenceInterval
- false
- hasNumericResultPublic
- false
- hasNumericResultPrivate
- false
- task_id
- 1071469
- sort_order
- DESCENDING
- source_name
- gpqa-main-zero-shot/task-068
- display_type
- percentages
- task_version
- 1
- task_display_type
- RAW_VALUES
- normalization_status
- normalized
- published_at
- 2026-10-21T18:15:21.342276500Z
- candidate_class
- modelVersion
- source_revision
- b32d007df5acbddafca23c7dd2beba8221c519cf7dd3445a1d7bc28740ffd0d8
- publisher_anchor
- openai/o4-mini-2025-04-16
- metric_population
- 26
- benchmark_evaluation
- true
- task_metadata_source
- benchmarkVersion
- candidate_metadata_source
- benchmarkVersionMappings
- evaluated at
- 2026-10-20T18:39:49.468Z
- source model id
- o4 mini
Original result ↗100%
- source id
- kaggleBenchmarks
- leaderboard name
- GPQA Diamond Zero‑Shot · gpqa-diamond-zero-shot/task-189
- dimension
- reasoning
- metric
- GPQA Diamond Zero‑Shot · gpqa-diamond-zero-shot/task-189
- rank
- 1
- aggregation scope
- kaggle
- true
- status
- published
- license
- Apache-2.0
- direction
- higherIsBetter
- conditions
- task
- id
- 984239
- name
- gpqa-diamond-zero-shot/task-189
- version
- 1
- sort_order
- DESCENDING
- display_type
- percentages
- benchmark
- id
- 7
- name
- GPQA Diamond Zero‑Shot
- type
- INDIVIDUAL
- owner
- open-benchmarks
- version
- 2
- categories
- general knowledge and reasoning
- science
- version_id
- 8176
- maintainer_tier
- Kaggle-maintained
- maintenance_level
- KAGGLE_MAINTAINED
- candidate
- candidate_class
- modelVersion
- record_kind
- aggregate
- attribution
- Kaggle · open-benchmarks · GPQA Diamond Zero‑Shot
- license_url
- https://www.apache.org/licenses/LICENSE-2.0
- metric_kind
- GPQA Diamond Zero‑Shot · gpqa-diamond-zero-shot/task-189
- rank_method
- competition rank among reported evaluations in identical conditions, including unlinked observations
- source_file
- 8176/leaderboard.json
- source_page
- https://www.kaggle.com/benchmarks/open-benchmarks/gpqa-zero-shot-diamond-set/versions/2
- measurements
- name
- gpqa-diamond-zero-shot/task-189
- unit
- percent
- scale
- 1
- metric
- GPQA Diamond Zero‑Shot · gpqa-diamond-zero-shot/task-189
- result
- resultCase
- numericResult
- booleanResult
- false
- numericResult
- value
- 1
- standardDeviation
- 0
- confidenceInterval
- 0
- hasStandardDeviation
- false
- hasConfidenceInterval
- false
- hasUnevenConfidenceInterval
- false
- taskVersionId
- 0
- evaluationDate
- 2026-10-12T18:15:38.5518115Z
- hasBooleanResult
- false
- hasNumericResult
- true
- hasTaskVersionId
- false
- hasEvaluationDate
- true
- numericResultNullable
- value
- 1
- standardDeviation
- 0
- confidenceInterval
- 0
- hasStandardDeviation
- false
- hasConfidenceInterval
- false
- hasUnevenConfidenceInterval
- false
- hasNumericResultPublic
- false
- hasNumericResultPrivate
- false
- task_id
- 984239
- sort_order
- DESCENDING
- source_name
- gpqa-diamond-zero-shot/task-189
- display_type
- percentages
- task_version
- 1
- task_display_type
- RAW_VALUES
- normalization_status
- normalized
- published_at
- 2026-10-12T17:49:58.132967500Z
- candidate_class
- modelVersion
- source_revision
- b32d007df5acbddafca23c7dd2beba8221c519cf7dd3445a1d7bc28740ffd0d8
- publisher_anchor
- openai/o4-mini-2025-04-16
- metric_population
- 32
- benchmark_evaluation
- true
- task_metadata_source
- benchmarkVersion
- candidate_metadata_source
- benchmarkVersionMappings
- evaluated at
- 2026-10-12T18:15:38.551Z
- source model id
- o4 mini
Original result ↗100%
- source id
- kaggleBenchmarks
- leaderboard name
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-176
- dimension
- reasoning
- metric
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-176
- rank
- 1
- aggregation scope
- kaggle
- true
- status
- published
- license
- Apache-2.0
- direction
- higherIsBetter
- conditions
- task
- id
- 1071590
- name
- gpqa-main-zero-shot/task-176
- version
- 1
- sort_order
- DESCENDING
- display_type
- percentages
- benchmark
- id
- 8
- name
- GPQA Main Zero‑Shot
- type
- INDIVIDUAL
- owner
- open-benchmarks
- version
- 2
- categories
- general knowledge and reasoning
- science
- version_id
- 8463
- maintainer_tier
- Kaggle-maintained
- maintenance_level
- KAGGLE_MAINTAINED
- candidate
- candidate_class
- modelVersion
- record_kind
- aggregate
- attribution
- Kaggle · open-benchmarks · GPQA Main Zero‑Shot
- license_url
- https://www.apache.org/licenses/LICENSE-2.0
- metric_kind
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-176
- rank_method
- competition rank among reported evaluations in identical conditions, including unlinked observations
- source_file
- 8463/leaderboard.json
- source_page
- https://www.kaggle.com/benchmarks/open-benchmarks/gpqa-zero-shot-main-set/versions/2
- measurements
- name
- gpqa-main-zero-shot/task-176
- unit
- percent
- scale
- 1
- metric
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-176
- result
- resultCase
- numericResult
- booleanResult
- false
- numericResult
- value
- 1
- standardDeviation
- 0
- confidenceInterval
- 0
- hasStandardDeviation
- false
- hasConfidenceInterval
- false
- hasUnevenConfidenceInterval
- false
- taskVersionId
- 0
- evaluationDate
- 2026-10-20T19:39:33.7664977Z
- hasBooleanResult
- false
- hasNumericResult
- true
- hasTaskVersionId
- false
- hasEvaluationDate
- true
- numericResultNullable
- value
- 1
- standardDeviation
- 0
- confidenceInterval
- 0
- hasStandardDeviation
- false
- hasConfidenceInterval
- false
- hasUnevenConfidenceInterval
- false
- hasNumericResultPublic
- false
- hasNumericResultPrivate
- false
- task_id
- 1071590
- sort_order
- DESCENDING
- source_name
- gpqa-main-zero-shot/task-176
- display_type
- percentages
- task_version
- 1
- task_display_type
- RAW_VALUES
- normalization_status
- normalized
- published_at
- 2026-10-21T18:15:21.342276500Z
- candidate_class
- modelVersion
- source_revision
- b32d007df5acbddafca23c7dd2beba8221c519cf7dd3445a1d7bc28740ffd0d8
- publisher_anchor
- openai/o4-mini-2025-04-16
- metric_population
- 26
- benchmark_evaluation
- true
- task_metadata_source
- benchmarkVersion
- candidate_metadata_source
- benchmarkVersionMappings
- evaluated at
- 2026-10-20T19:39:33.766Z
- source model id
- o4 mini
Original result ↗100%
- source id
- kaggleBenchmarks
- leaderboard name
- GPQA Diamond Zero‑Shot · gpqa-diamond-zero-shot/task-095
- dimension
- reasoning
- metric
- GPQA Diamond Zero‑Shot · gpqa-diamond-zero-shot/task-095
- rank
- 1
- aggregation scope
- kaggle
- true
- status
- published
- license
- Apache-2.0
- direction
- higherIsBetter
- conditions
- task
- id
- 984144
- name
- gpqa-diamond-zero-shot/task-095
- version
- 1
- sort_order
- DESCENDING
- display_type
- percentages
- benchmark
- id
- 7
- name
- GPQA Diamond Zero‑Shot
- type
- INDIVIDUAL
- owner
- open-benchmarks
- version
- 2
- categories
- general knowledge and reasoning
- science
- version_id
- 8176
- maintainer_tier
- Kaggle-maintained
- maintenance_level
- KAGGLE_MAINTAINED
- candidate
- candidate_class
- modelVersion
- record_kind
- aggregate
- attribution
- Kaggle · open-benchmarks · GPQA Diamond Zero‑Shot
- license_url
- https://www.apache.org/licenses/LICENSE-2.0
- metric_kind
- GPQA Diamond Zero‑Shot · gpqa-diamond-zero-shot/task-095
- rank_method
- competition rank among reported evaluations in identical conditions, including unlinked observations
- source_file
- 8176/leaderboard.json
- source_page
- https://www.kaggle.com/benchmarks/open-benchmarks/gpqa-zero-shot-diamond-set/versions/2
- measurements
- name
- gpqa-diamond-zero-shot/task-095
- unit
- percent
- scale
- 1
- metric
- GPQA Diamond Zero‑Shot · gpqa-diamond-zero-shot/task-095
- result
- resultCase
- numericResult
- booleanResult
- false
- numericResult
- value
- 1
- standardDeviation
- 0
- confidenceInterval
- 0
- hasStandardDeviation
- false
- hasConfidenceInterval
- false
- hasUnevenConfidenceInterval
- false
- taskVersionId
- 0
- evaluationDate
- 2026-10-12T18:12:19.4136386Z
- hasBooleanResult
- false
- hasNumericResult
- true
- hasTaskVersionId
- false
- hasEvaluationDate
- true
- numericResultNullable
- value
- 1
- standardDeviation
- 0
- confidenceInterval
- 0
- hasStandardDeviation
- false
- hasConfidenceInterval
- false
- hasUnevenConfidenceInterval
- false
- hasNumericResultPublic
- false
- hasNumericResultPrivate
- false
- task_id
- 984144
- sort_order
- DESCENDING
- source_name
- gpqa-diamond-zero-shot/task-095
- display_type
- percentages
- task_version
- 1
- task_display_type
- RAW_VALUES
- normalization_status
- normalized
- published_at
- 2026-10-12T17:49:58.132967500Z
- candidate_class
- modelVersion
- source_revision
- b32d007df5acbddafca23c7dd2beba8221c519cf7dd3445a1d7bc28740ffd0d8
- publisher_anchor
- openai/o4-mini-2025-04-16
- metric_population
- 32
- benchmark_evaluation
- true
- task_metadata_source
- benchmarkVersion
- candidate_metadata_source
- benchmarkVersionMappings
- evaluated at
- 2026-10-12T18:12:19.413Z
- source model id
- o4 mini
Original result ↗100%
- source id
- kaggleBenchmarks
- leaderboard name
- GPQA Diamond Zero‑Shot · gpqa-diamond-zero-shot/task-057
- dimension
- reasoning
- metric
- GPQA Diamond Zero‑Shot · gpqa-diamond-zero-shot/task-057
- rank
- 1
- aggregation scope
- kaggle
- true
- status
- published
- license
- Apache-2.0
- direction
- higherIsBetter
- conditions
- task
- id
- 984106
- name
- gpqa-diamond-zero-shot/task-057
- version
- 1
- sort_order
- DESCENDING
- display_type
- percentages
- benchmark
- id
- 7
- name
- GPQA Diamond Zero‑Shot
- type
- INDIVIDUAL
- owner
- open-benchmarks
- version
- 2
- categories
- general knowledge and reasoning
- science
- version_id
- 8176
- maintainer_tier
- Kaggle-maintained
- maintenance_level
- KAGGLE_MAINTAINED
- candidate
- candidate_class
- modelVersion
- record_kind
- aggregate
- attribution
- Kaggle · open-benchmarks · GPQA Diamond Zero‑Shot
- license_url
- https://www.apache.org/licenses/LICENSE-2.0
- metric_kind
- GPQA Diamond Zero‑Shot · gpqa-diamond-zero-shot/task-057
- rank_method
- competition rank among reported evaluations in identical conditions, including unlinked observations
- source_file
- 8176/leaderboard.json
- source_page
- https://www.kaggle.com/benchmarks/open-benchmarks/gpqa-zero-shot-diamond-set/versions/2
- measurements
- name
- gpqa-diamond-zero-shot/task-057
- unit
- percent
- scale
- 1
- metric
- GPQA Diamond Zero‑Shot · gpqa-diamond-zero-shot/task-057
- result
- resultCase
- numericResult
- booleanResult
- false
- numericResult
- value
- 1
- standardDeviation
- 0
- confidenceInterval
- 0
- hasStandardDeviation
- false
- hasConfidenceInterval
- false
- hasUnevenConfidenceInterval
- false
- taskVersionId
- 0
- evaluationDate
- 2026-10-12T18:11:58.4966747Z
- hasBooleanResult
- false
- hasNumericResult
- true
- hasTaskVersionId
- false
- hasEvaluationDate
- true
- numericResultNullable
- value
- 1
- standardDeviation
- 0
- confidenceInterval
- 0
- hasStandardDeviation
- false
- hasConfidenceInterval
- false
- hasUnevenConfidenceInterval
- false
- hasNumericResultPublic
- false
- hasNumericResultPrivate
- false
- task_id
- 984106
- sort_order
- DESCENDING
- source_name
- gpqa-diamond-zero-shot/task-057
- display_type
- percentages
- task_version
- 1
- task_display_type
- RAW_VALUES
- normalization_status
- normalized
- published_at
- 2026-10-12T17:49:58.132967500Z
- candidate_class
- modelVersion
- source_revision
- b32d007df5acbddafca23c7dd2beba8221c519cf7dd3445a1d7bc28740ffd0d8
- publisher_anchor
- openai/o4-mini-2025-04-16
- metric_population
- 32
- benchmark_evaluation
- true
- task_metadata_source
- benchmarkVersion
- candidate_metadata_source
- benchmarkVersionMappings
- evaluated at
- 2026-10-12T18:11:58.496Z
- source model id
- o4 mini
Original result ↗100%
- source id
- kaggleBenchmarks
- leaderboard name
- GPQA Diamond Zero‑Shot · gpqa-diamond-zero-shot/task-050
- dimension
- reasoning
- metric
- GPQA Diamond Zero‑Shot · gpqa-diamond-zero-shot/task-050
- rank
- 1
- aggregation scope
- kaggle
- true
- status
- published
- license
- Apache-2.0
- direction
- higherIsBetter
- conditions
- task
- id
- 984099
- name
- gpqa-diamond-zero-shot/task-050
- version
- 1
- sort_order
- DESCENDING
- display_type
- percentages
- benchmark
- id
- 7
- name
- GPQA Diamond Zero‑Shot
- type
- INDIVIDUAL
- owner
- open-benchmarks
- version
- 2
- categories
- general knowledge and reasoning
- science
- version_id
- 8176
- maintainer_tier
- Kaggle-maintained
- maintenance_level
- KAGGLE_MAINTAINED
- candidate
- candidate_class
- modelVersion
- record_kind
- aggregate
- attribution
- Kaggle · open-benchmarks · GPQA Diamond Zero‑Shot
- license_url
- https://www.apache.org/licenses/LICENSE-2.0
- metric_kind
- GPQA Diamond Zero‑Shot · gpqa-diamond-zero-shot/task-050
- rank_method
- competition rank among reported evaluations in identical conditions, including unlinked observations
- source_file
- 8176/leaderboard.json
- source_page
- https://www.kaggle.com/benchmarks/open-benchmarks/gpqa-zero-shot-diamond-set/versions/2
- measurements
- name
- gpqa-diamond-zero-shot/task-050
- unit
- percent
- scale
- 1
- metric
- GPQA Diamond Zero‑Shot · gpqa-diamond-zero-shot/task-050
- result
- resultCase
- numericResult
- booleanResult
- false
- numericResult
- value
- 1
- standardDeviation
- 0
- confidenceInterval
- 0
- hasStandardDeviation
- false
- hasConfidenceInterval
- false
- hasUnevenConfidenceInterval
- false
- taskVersionId
- 0
- evaluationDate
- 2026-10-12T18:11:56.6391813Z
- hasBooleanResult
- false
- hasNumericResult
- true
- hasTaskVersionId
- false
- hasEvaluationDate
- true
- numericResultNullable
- value
- 1
- standardDeviation
- 0
- confidenceInterval
- 0
- hasStandardDeviation
- false
- hasConfidenceInterval
- false
- hasUnevenConfidenceInterval
- false
- hasNumericResultPublic
- false
- hasNumericResultPrivate
- false
- task_id
- 984099
- sort_order
- DESCENDING
- source_name
- gpqa-diamond-zero-shot/task-050
- display_type
- percentages
- task_version
- 1
- task_display_type
- RAW_VALUES
- normalization_status
- normalized
- published_at
- 2026-10-12T17:49:58.132967500Z
- candidate_class
- modelVersion
- source_revision
- b32d007df5acbddafca23c7dd2beba8221c519cf7dd3445a1d7bc28740ffd0d8
- publisher_anchor
- openai/o4-mini-2025-04-16
- metric_population
- 32
- benchmark_evaluation
- true
- task_metadata_source
- benchmarkVersion
- candidate_metadata_source
- benchmarkVersionMappings
- evaluated at
- 2026-10-12T18:11:56.639Z
- source model id
- o4 mini
Original result ↗100%
- source id
- kaggleBenchmarks
- leaderboard name
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-212
- dimension
- reasoning
- metric
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-212
- rank
- 1
- aggregation scope
- kaggle
- true
- status
- published
- license
- Apache-2.0
- direction
- higherIsBetter
- conditions
- task
- id
- 1071630
- name
- gpqa-main-zero-shot/task-212
- version
- 1
- sort_order
- DESCENDING
- display_type
- percentages
- benchmark
- id
- 8
- name
- GPQA Main Zero‑Shot
- type
- INDIVIDUAL
- owner
- open-benchmarks
- version
- 2
- categories
- general knowledge and reasoning
- science
- version_id
- 8463
- maintainer_tier
- Kaggle-maintained
- maintenance_level
- KAGGLE_MAINTAINED
- candidate
- candidate_class
- modelVersion
- record_kind
- aggregate
- attribution
- Kaggle · open-benchmarks · GPQA Main Zero‑Shot
- license_url
- https://www.apache.org/licenses/LICENSE-2.0
- metric_kind
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-212
- rank_method
- competition rank among reported evaluations in identical conditions, including unlinked observations
- source_file
- 8463/leaderboard.json
- source_page
- https://www.kaggle.com/benchmarks/open-benchmarks/gpqa-zero-shot-main-set/versions/2
- measurements
- name
- gpqa-main-zero-shot/task-212
- unit
- percent
- scale
- 1
- metric
- GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-212
- result
- resultCase
- numericResult
- booleanResult
- false
- numericResult
- value
- 1
- standardDeviation
- 0
- confidenceInterval
- 0
- hasStandardDeviation
- false
- hasConfidenceInterval
- false
- hasUnevenConfidenceInterval
- false
- taskVersionId
- 0
- evaluationDate
- 2026-10-20T19:56:07.5654273Z
- hasBooleanResult
- false
- hasNumericResult
- true
- hasTaskVersionId
- false
- hasEvaluationDate
- true
- numericResultNullable
- value
- 1
- standardDeviation
- 0
- confidenceInterval
- 0
- hasStandardDeviation
- false
- hasConfidenceInterval
- false
- hasUnevenConfidenceInterval
- false
- hasNumericResultPublic
- false
- hasNumericResultPrivate
- false
- task_id
- 1071630
- sort_order
- DESCENDING
- source_name
- gpqa-main-zero-shot/task-212
- display_type
- percentages
- task_version
- 1
- task_display_type
- RAW_VALUES
- normalization_status
- normalized
- published_at
- 2026-10-21T18:15:21.342276500Z
- candidate_class
- modelVersion
- source_revision
- b32d007df5acbddafca23c7dd2beba8221c519cf7dd3445a1d7bc28740ffd0d8
- publisher_anchor
- openai/o4-mini-2025-04-16
- metric_population
- 26
- benchmark_evaluation
- true
- task_metadata_source
- benchmarkVersion
- candidate_metadata_source
- benchmarkVersionMappings
- evaluated at
- 2026-10-20T19:56:07.565Z
- source model id
- o4 mini
Original result ↗Published prices
1–15 / 15
$0.28
- Inference provider
- cloudflareWorkersAI
- Platform
- OpenAI
- Unit
- cache_read_tokens_per_million
- Currency
- USD
- Billing label
- Cached input tokens (per 1M)
- Pricing via
- Cloudflare AI Gateway
$1.10
- Inference provider
- cloudflareWorkersAI
- Platform
- OpenAI
- Unit
- input_tokens_per_million
- Currency
- USD
- Billing label
- Input tokens (per 1M)
- Pricing via
- Cloudflare AI Gateway
$4.40
- Inference provider
- cloudflareWorkersAI
- Platform
- OpenAI
- Unit
- output_tokens_per_million
- Currency
- USD
- Billing label
- Output tokens (per 1M)
- Pricing via
- Cloudflare AI Gateway
$0.14
- Inference provider
- openRouter
- Platform
- OpenRouter
- Unit
- cache_read_tokens_per_million
- Currency
- USD
- Billing conditions
- mode
- batch
$0.55
- Inference provider
- openRouter
- Platform
- OpenRouter
- Unit
- input_tokens_per_million
- Currency
- USD
- Billing conditions
- mode
- batch
$2.20
- Inference provider
- openRouter
- Platform
- OpenRouter
- Unit
- output_tokens_per_million
- Currency
- USD
- Billing conditions
- mode
- batch
$0.01
- Inference provider
- openRouter
- Platform
- OpenRouter
- Unit
- web_search
- Currency
- USD
- Billing conditions
- mode
- batch
$0.28
- Inference provider
- vercelAIGatewayCatalog
- Platform
- azure
- Unit
- cache_read_tokens_per_million
- Currency
- USD
- Billing conditions
- endpoint
- azure | openai/o4-mini
- model_type
- language
- billing_scope
- via Vercel AI Gateway
- Billing label
- Cached input / 1M
- Pricing via
- Vercel AI Gateway
$1.10
- Inference provider
- vercelAIGatewayCatalog
- Platform
- azure
- Unit
- input_tokens_per_million
- Currency
- USD
- Billing conditions
- endpoint
- azure | openai/o4-mini
- model_type
- language
- billing_scope
- via Vercel AI Gateway
- Billing label
- Input tokens / 1M
- Pricing via
- Vercel AI Gateway
$4.40
- Inference provider
- vercelAIGatewayCatalog
- Platform
- azure
- Unit
- output_tokens_per_million
- Currency
- USD
- Billing conditions
- endpoint
- azure | openai/o4-mini
- model_type
- language
- billing_scope
- via Vercel AI Gateway
- Billing label
- Output tokens / 1M
- Pricing via
- Vercel AI Gateway
$0.01
- Inference provider
- vercelAIGatewayCatalog
- Platform
- azure
- Unit
- request
- Currency
- USD
- Billing conditions
- charge
- web_search
- endpoint
- azure | openai/o4-mini
- model_type
- language
- billing_scope
- via Vercel AI Gateway
- Billing label
- Web search
- Pricing via
- Vercel AI Gateway
$0.28
- Inference provider
- vercelAIGatewayCatalog
- Platform
- Vercel AI Gateway
- Unit
- cache_read_tokens_per_million
- Currency
- USD
- Billing conditions
- model_type
- language
- quote_scope
- gateway_catalog
- billing_scope
- Vercel public catalog; endpoint prices may differ
- Billing label
- Cached input / 1M
- Pricing via
- Vercel AI Gateway
$1.10
- Inference provider
- vercelAIGatewayCatalog
- Platform
- Vercel AI Gateway
- Unit
- input_tokens_per_million
- Currency
- USD
- Billing conditions
- model_type
- language
- quote_scope
- gateway_catalog
- billing_scope
- Vercel public catalog; endpoint prices may differ
- Billing label
- Input tokens / 1M
- Pricing via
- Vercel AI Gateway
$4.40
- Inference provider
- vercelAIGatewayCatalog
- Platform
- Vercel AI Gateway
- Unit
- output_tokens_per_million
- Currency
- USD
- Billing conditions
- model_type
- language
- quote_scope
- gateway_catalog
- billing_scope
- Vercel public catalog; endpoint prices may differ
- Billing label
- Output tokens / 1M
- Pricing via
- Vercel AI Gateway
$0.01
- Inference provider
- vercelAIGatewayCatalog
- Platform
- Vercel AI Gateway
- Unit
- request
- Currency
- USD
- Billing conditions
- charge
- web_search
- model_type
- language
- quote_scope
- gateway_catalog
- billing_scope
- Vercel public catalog; endpoint prices may differ
- Billing label
- Web search
- Pricing via
- Vercel AI Gateway
Provider specifications & access
1–3 / 3
- source model id
- azure-openai/o4-mini@2025-04-16
- hosting
- foundry-managed-compute
- license
- custom
- context length
- 200000
- max output tokens
- 100000
- source model id
- openai/o4-mini
- hosting
- third-party
- context length
- 200000
- max output tokens
- 100000
- source model id
- openai/o4-mini
- hosting
- vercel-gateway
- context length
- 200000
- max output tokens
- 100000
- capabilities
- promptCaching
- reasoning
- structuredOutputs
- toolCalling
- visionInput
- webSearch
Safety assessments
1–1 / 1
35 / 100
- source id
- stanfordFMTI
- canonical model id
- openai/o4-mini-2025-04-16
- taxonomy
- transparency
- hazard category
- transparency
- hazard name
- Organization transparency
- policy area
- OpenAI
- assessment type
- organization-level disclosure
- metric
- disclosure_score
- result label
- 35 / 100
- sample size
- 100
- updated at
- 2026-11-09T18:08:05.227Z
Source specification observations
1–11 / 11
- field
- context_length
- value
- 200000
- normalized value
- 200000
- scope
- source-model
- source id
- azureFoundry
- confidence
- 1
- field
- context_length
- value
- 200000
- normalized value
- 200000
- scope
- source-model
- source id
- cloudflareWorkersAI
- confidence
- 1
- field
- context_length
- value
- 200000
- normalized value
- 200000
- scope
- source-model
- source id
- openRouter
- confidence
- 1
- field
- context_length
- value
- 200000
- normalized value
- 200000
- scope
- source-model
- source id
- vercelAIGatewayCatalog
- confidence
- 1
- field
- knowledge_cutoff
- reported field
- knowledge_cutoff
- value
- 2024-06-30
- normalized value
- 2024-06-30
- precision
- day
- scope
- source-model
- source id
- openRouter
- confidence
- 1
- field
- knowledge_cutoff
- reported field
- knowledge_cutoff
- value
- 2024-06-01
- normalized value
- 2024-06-01
- precision
- day
- scope
- source-model
- source id
- vercelAIGatewayCatalog
- confidence
- 1
- field
- max_output_tokens
- value
- 100000
- normalized value
- 100000
- scope
- source-model
- source id
- azureFoundry
- confidence
- 1
- field
- max_output_tokens
- value
- 100000
- normalized value
- 100000
- scope
- source-model
- source id
- cloudflareWorkersAI
- confidence
- 1
- field
- max_output_tokens
- value
- 100000
- normalized value
- 100000
- scope
- source-model
- source id
- openRouter
- confidence
- 1
- field
- max_output_tokens
- value
- 100000
- normalized value
- 100000
- scope
- source-model
- source id
- vercelAIGatewayCatalog
- confidence
- 1
- field
- knowledge_cutoff
- reported field
- training_data_cutoff
- value
- May 2024
- normalized value
- 2024-05
- precision
- month
- scope
- source-model
- source id
- azureFoundry
- confidence
- 1
Explore more models