Qwen: Qwen2.5 VL 72B Instruct
Qwen2.5-VL is proficient in recognizing common objects such as flowers, birds, fish, and insects. It is also highly capable of analyzing texts, charts, icons, graphics, and layouts within images.
qwen/qwen2.5-vl-72b-instruct
Qwen2.5-VL is proficient in recognizing common objects such as flowers, birds, fish, and insects. It is also highly capable of analyzing texts, charts, icons, graphics, and layouts within images.
Specifications & capabilities
- Input
- text, image
- Output
- text
- Context window · tokens
- 128000
- Knowledge cutoff
- 2024-06-30
- Published model ID
- Qwen2.5-VL-72B-Instruct
- Capabilities
- multimodal, promptCaching, structuredOutputs, visionInput
Evaluations
1–20 / 131
100%
- source id
- mmBench
- leaderboard name
- ocr
- dimension
- vision
- metric
- ocr
- rank
- 1
- aggregation scope
- status
- published
- direction
- higherIsBetter
- conditions
- suite
- MMBench_TEST_CN_Data
- verified
- Yes
- metric_kind
- ocr
- rank_method
- competition rank among reported evaluations in identical conditions, including unlinked observations
- source_file
- mmbench-data.json
- source_revision
- acb08c72ac038d24cf7785254678efe725354fe91091bfa10a667832ad964b24
- metric_population
- 163
- benchmark_evaluation
- true
- source model id
- Qwen2.5-VL-72B
Original result ↗100%
- source id
- mmBench
- leaderboard name
- ocr
- dimension
- vision
- metric
- ocr
- rank
- 1
- aggregation scope
- status
- published
- direction
- higherIsBetter
- conditions
- suite
- MMBench_TEST_EN_V11_Data
- verified
- Yes
- metric_kind
- ocr
- rank_method
- competition rank among reported evaluations in identical conditions, including unlinked observations
- source_file
- mmbench-data.json
- source_revision
- acb08c72ac038d24cf7785254678efe725354fe91091bfa10a667832ad964b24
- metric_population
- 295
- benchmark_evaluation
- true
- source model id
- Qwen2.5-VL-72B
Original result ↗100%
- source id
- mmBench
- leaderboard name
- ocr
- dimension
- vision
- metric
- ocr
- rank
- 1
- aggregation scope
- status
- published
- direction
- higherIsBetter
- conditions
- suite
- MMBench_TEST_CN_V11_Data
- verified
- Yes
- metric_kind
- ocr
- rank_method
- competition rank among reported evaluations in identical conditions, including unlinked observations
- source_file
- mmbench-data.json
- source_revision
- acb08c72ac038d24cf7785254678efe725354fe91091bfa10a667832ad964b24
- metric_population
- 295
- benchmark_evaluation
- true
- source model id
- Qwen2.5-VL-72B
Original result ↗100%
- source id
- mmBench
- leaderboard name
- ocr
- dimension
- vision
- metric
- ocr
- rank
- 1
- aggregation scope
- status
- published
- direction
- higherIsBetter
- conditions
- suite
- MMBench_TEST_EN_Data
- verified
- Yes
- metric_kind
- ocr
- rank_method
- competition rank among reported evaluations in identical conditions, including unlinked observations
- source_file
- mmbench-data.json
- source_revision
- acb08c72ac038d24cf7785254678efe725354fe91091bfa10a667832ad964b24
- metric_population
- 162
- benchmark_evaluation
- true
- source model id
- Qwen2.5-VL-72B
Original result ↗98.9%
- source id
- mmBench
- leaderboard name
- image_topic
- dimension
- vision
- metric
- image_topic
- rank
- 1
- aggregation scope
- status
- published
- direction
- higherIsBetter
- conditions
- suite
- MMBench_TEST_CN_V11_Data
- verified
- Yes
- metric_kind
- image_topic
- rank_method
- competition rank among reported evaluations in identical conditions, including unlinked observations
- source_file
- mmbench-data.json
- source_revision
- acb08c72ac038d24cf7785254678efe725354fe91091bfa10a667832ad964b24
- metric_population
- 295
- benchmark_evaluation
- true
- source model id
- Qwen2.5-VL-72B
Original result ↗96.3%
- source id
- mmBench
- leaderboard name
- FP-S
- dimension
- vision
- metric
- FP-S
- rank
- 1
- aggregation scope
- status
- published
- direction
- higherIsBetter
- conditions
- suite
- MMBench_TEST_CN_V11_Data
- verified
- Yes
- metric_kind
- FP-S
- rank_method
- competition rank among reported evaluations in identical conditions, including unlinked observations
- source_file
- mmbench-data.json
- source_revision
- acb08c72ac038d24cf7785254678efe725354fe91091bfa10a667832ad964b24
- metric_population
- 295
- benchmark_evaluation
- true
- source model id
- Qwen2.5-VL-72B
Original result ↗96.2%
- source id
- mmBench
- leaderboard name
- FP-S
- dimension
- vision
- metric
- FP-S
- rank
- 1
- aggregation scope
- status
- published
- direction
- higherIsBetter
- conditions
- suite
- MMBench_TEST_CN_Data
- verified
- Yes
- metric_kind
- FP-S
- rank_method
- competition rank among reported evaluations in identical conditions, including unlinked observations
- source_file
- mmbench-data.json
- source_revision
- acb08c72ac038d24cf7785254678efe725354fe91091bfa10a667832ad964b24
- metric_population
- 236
- benchmark_evaluation
- true
- source model id
- Qwen2.5-VL-72B
Original result ↗95.2%
- source id
- mmBench
- leaderboard name
- FP-S
- dimension
- vision
- metric
- FP-S
- rank
- 1
- aggregation scope
- status
- published
- direction
- higherIsBetter
- conditions
- suite
- MMBench_TEST_EN_Data
- verified
- Yes
- metric_kind
- FP-S
- rank_method
- competition rank among reported evaluations in identical conditions, including unlinked observations
- source_file
- mmbench-data.json
- source_revision
- acb08c72ac038d24cf7785254678efe725354fe91091bfa10a667832ad964b24
- metric_population
- 235
- benchmark_evaluation
- true
- source model id
- Qwen2.5-VL-72B
Original result ↗93.8%
- source id
- mmBench
- leaderboard name
- image_scene
- dimension
- vision
- metric
- image_scene
- rank
- 1
- aggregation scope
- status
- published
- direction
- higherIsBetter
- conditions
- suite
- MMBench_TEST_EN_V11_Data
- verified
- Yes
- metric_kind
- image_scene
- rank_method
- competition rank among reported evaluations in identical conditions, including unlinked observations
- source_file
- mmbench-data.json
- source_revision
- acb08c72ac038d24cf7785254678efe725354fe91091bfa10a667832ad964b24
- metric_population
- 295
- benchmark_evaluation
- true
- source model id
- Qwen2.5-VL-72B
Original result ↗0.937
- source id
- llmStatsArchive
- leaderboard name
- Android Control Low_EM · Normalized (reported)
- dimension
- general
- metric
- Android Control Low_EM · Normalized (reported)
- rank
- 1
- aggregation scope
- status
- archived
- license
- CC-BY-4.0
- direction
- higherIsBetter
- conditions
- language
- en
- benchmark
- name
- Android Control Low_EM
- language
- en
- modality
- multimodal
- verified
- false
- max_score
- 1
- created_at
- 2025-07-19T19:56:14.800337+00:00
- updated_at
- 2025-07-19T19:56:14.800337+00:00
- description
- Android control benchmark evaluating autonomous agents on mobile device interaction tasks with low exact match scoring criteria
- benchmark_id
- android-control-low-em
- multilingual
- false
- analysis_method
- EM
- is_self_reported
- true
- verified_by_llmstats
- false
- attribution
- Jonathan Chavez Tamales and LLM Stats contributors (copyright 2024 jc); TensorTrace structures original results and derives scoped ranks.
- license_url
- https://creativecommons.org/licenses/by/4.0/
- metric_kind
- Android Control Low_EM · Normalized (reported)
- rank_method
- competition rank among reported evaluations in identical conditions, including unlinked observations
- source_file
- data/organizations/qwen/models/qwen2.5-vl-72b/benchmarks.json
- archive_kind
- llmStats
- evidence_url
- https://huggingface.co/Qwen/Qwen2.5-VL-72B-Instruct
- git_revision
- 872b75f63b8d853c46ed17856424029a0c47b1d2
- published_at
- 2025-10-24T17:47:59Z
- snapshot_date
- 2025-10-24T17:47:59Z
- source_revision
- 6e27aa41f51fa8cb327602349240077d1c7acae9fead8ba2f07db0d549927ddc
- publisher_anchor
- qwen/qwen2.5-vl-72b
- metric_population
- 3
- current_catalog_url
- https://llm-stats.com/
- benchmark_evaluation
- true
- source model id
- qwen2.5-vl-72b
Original result ↗0.937
- source id
- llmStatsArchive
- leaderboard name
- Android Control Low_EM · Raw
- dimension
- general
- metric
- Android Control Low_EM · Raw
- rank
- 1
- aggregation scope
- status
- archived
- license
- CC-BY-4.0
- direction
- higherIsBetter
- conditions
- language
- en
- benchmark
- name
- Android Control Low_EM
- language
- en
- modality
- multimodal
- verified
- false
- max_score
- 1
- created_at
- 2025-07-19T19:56:14.800337+00:00
- updated_at
- 2025-07-19T19:56:14.800337+00:00
- description
- Android control benchmark evaluating autonomous agents on mobile device interaction tasks with low exact match scoring criteria
- benchmark_id
- android-control-low-em
- multilingual
- false
- analysis_method
- EM
- is_self_reported
- true
- verified_by_llmstats
- false
- attribution
- Jonathan Chavez Tamales and LLM Stats contributors (copyright 2024 jc); TensorTrace structures original results and derives scoped ranks.
- license_url
- https://creativecommons.org/licenses/by/4.0/
- metric_kind
- Android Control Low_EM · Raw
- rank_method
- competition rank among reported evaluations in identical conditions, including unlinked observations
- source_file
- data/organizations/qwen/models/qwen2.5-vl-72b/benchmarks.json
- archive_kind
- llmStats
- evidence_url
- https://huggingface.co/Qwen/Qwen2.5-VL-72B-Instruct
- git_revision
- 872b75f63b8d853c46ed17856424029a0c47b1d2
- published_at
- 2025-10-24T17:47:59Z
- snapshot_date
- 2025-10-24T17:47:59Z
- source_revision
- 6e27aa41f51fa8cb327602349240077d1c7acae9fead8ba2f07db0d549927ddc
- publisher_anchor
- qwen/qwen2.5-vl-72b
- metric_population
- 3
- current_catalog_url
- https://llm-stats.com/
- benchmark_evaluation
- true
- source model id
- qwen2.5-vl-72b
Original result ↗91.3%
- source id
- mmBench
- leaderboard name
- object_localization
- dimension
- vision
- metric
- object_localization
- rank
- 1
- aggregation scope
- status
- published
- direction
- higherIsBetter
- conditions
- suite
- MMBench_TEST_CN_Data
- verified
- Yes
- metric_kind
- object_localization
- rank_method
- competition rank among reported evaluations in identical conditions, including unlinked observations
- source_file
- mmbench-data.json
- source_revision
- acb08c72ac038d24cf7785254678efe725354fe91091bfa10a667832ad964b24
- metric_population
- 163
- benchmark_evaluation
- true
- source model id
- Qwen2.5-VL-72B
Original result ↗90.4%
- source id
- mmBench
- leaderboard name
- object_localization
- dimension
- vision
- metric
- object_localization
- rank
- 1
- aggregation scope
- status
- published
- direction
- higherIsBetter
- conditions
- suite
- MMBench_TEST_EN_Data
- verified
- Yes
- metric_kind
- object_localization
- rank_method
- competition rank among reported evaluations in identical conditions, including unlinked observations
- source_file
- mmbench-data.json
- source_revision
- acb08c72ac038d24cf7785254678efe725354fe91091bfa10a667832ad964b24
- metric_population
- 162
- benchmark_evaluation
- true
- source model id
- Qwen2.5-VL-72B
Original result ↗89.1%
- source id
- mmBench
- leaderboard name
- CP
- dimension
- vision
- metric
- CP
- rank
- 1
- aggregation scope
- status
- published
- direction
- higherIsBetter
- conditions
- suite
- MMBench_TEST_EN_Data
- verified
- Yes
- metric_kind
- CP
- rank_method
- competition rank among reported evaluations in identical conditions, including unlinked observations
- source_file
- mmbench-data.json
- source_revision
- acb08c72ac038d24cf7785254678efe725354fe91091bfa10a667832ad964b24
- metric_population
- 235
- benchmark_evaluation
- true
- source model id
- Qwen2.5-VL-72B
Original result ↗0.832
- source id
- llmStatsArchive
- leaderboard name
- AITZ_EM · Normalized (reported)
- dimension
- general
- metric
- AITZ_EM · Normalized (reported)
- rank
- 1
- aggregation scope
- status
- archived
- license
- CC-BY-4.0
- direction
- higherIsBetter
- conditions
- language
- en
- benchmark
- name
- AITZ_EM
- language
- en
- modality
- multimodal
- verified
- false
- max_score
- 1
- created_at
- 2025-07-19T19:56:14.785085+00:00
- paper_link
- https://arxiv.org/abs/2403.02713
- updated_at
- 2025-07-19T19:56:14.785085+00:00
- description
- Android-In-The-Zoo (AitZ) benchmark for evaluating autonomous GUI agents on smartphones. Contains 18,643 screen-action pairs with chain-of-action-thought annotations spanning over 70 Android apps. Designed to connect perception (screen layouts and UI elements) with cognition (action decision-making) for natural language-triggered smartphone task completion.
- benchmark_id
- aitz-em
- multilingual
- false
- analysis_method
- EM
- is_self_reported
- true
- verified_by_llmstats
- false
- attribution
- Jonathan Chavez Tamales and LLM Stats contributors (copyright 2024 jc); TensorTrace structures original results and derives scoped ranks.
- license_url
- https://creativecommons.org/licenses/by/4.0/
- metric_kind
- AITZ_EM · Normalized (reported)
- rank_method
- competition rank among reported evaluations in identical conditions, including unlinked observations
- source_file
- data/organizations/qwen/models/qwen2.5-vl-72b/benchmarks.json
- archive_kind
- llmStats
- evidence_url
- https://huggingface.co/Qwen/Qwen2.5-VL-72B-Instruct
- git_revision
- 872b75f63b8d853c46ed17856424029a0c47b1d2
- published_at
- 2025-10-24T17:47:59Z
- snapshot_date
- 2025-10-24T17:47:59Z
- source_revision
- 6e27aa41f51fa8cb327602349240077d1c7acae9fead8ba2f07db0d549927ddc
- publisher_anchor
- qwen/qwen2.5-vl-72b
- metric_population
- 3
- current_catalog_url
- https://llm-stats.com/
- benchmark_evaluation
- true
- source model id
- qwen2.5-vl-72b
Original result ↗0.832
- source id
- llmStatsArchive
- leaderboard name
- AITZ_EM · Raw
- dimension
- general
- metric
- AITZ_EM · Raw
- rank
- 1
- aggregation scope
- status
- archived
- license
- CC-BY-4.0
- direction
- higherIsBetter
- conditions
- language
- en
- benchmark
- name
- AITZ_EM
- language
- en
- modality
- multimodal
- verified
- false
- max_score
- 1
- created_at
- 2025-07-19T19:56:14.785085+00:00
- paper_link
- https://arxiv.org/abs/2403.02713
- updated_at
- 2025-07-19T19:56:14.785085+00:00
- description
- Android-In-The-Zoo (AitZ) benchmark for evaluating autonomous GUI agents on smartphones. Contains 18,643 screen-action pairs with chain-of-action-thought annotations spanning over 70 Android apps. Designed to connect perception (screen layouts and UI elements) with cognition (action decision-making) for natural language-triggered smartphone task completion.
- benchmark_id
- aitz-em
- multilingual
- false
- analysis_method
- EM
- is_self_reported
- true
- verified_by_llmstats
- false
- attribution
- Jonathan Chavez Tamales and LLM Stats contributors (copyright 2024 jc); TensorTrace structures original results and derives scoped ranks.
- license_url
- https://creativecommons.org/licenses/by/4.0/
- metric_kind
- AITZ_EM · Raw
- rank_method
- competition rank among reported evaluations in identical conditions, including unlinked observations
- source_file
- data/organizations/qwen/models/qwen2.5-vl-72b/benchmarks.json
- archive_kind
- llmStats
- evidence_url
- https://huggingface.co/Qwen/Qwen2.5-VL-72B-Instruct
- git_revision
- 872b75f63b8d853c46ed17856424029a0c47b1d2
- published_at
- 2025-10-24T17:47:59Z
- snapshot_date
- 2025-10-24T17:47:59Z
- source_revision
- 6e27aa41f51fa8cb327602349240077d1c7acae9fead8ba2f07db0d549927ddc
- publisher_anchor
- qwen/qwen2.5-vl-72b
- metric_population
- 3
- current_catalog_url
- https://llm-stats.com/
- benchmark_evaluation
- true
- source model id
- qwen2.5-vl-72b
Original result ↗98.9%
- source id
- mmBench
- leaderboard name
- image_topic
- dimension
- vision
- metric
- image_topic
- rank
- 2
- aggregation scope
- status
- published
- direction
- higherIsBetter
- conditions
- suite
- MMBench_TEST_EN_V11_Data
- verified
- Yes
- metric_kind
- image_topic
- rank_method
- competition rank among reported evaluations in identical conditions, including unlinked observations
- source_file
- mmbench-data.json
- source_revision
- acb08c72ac038d24cf7785254678efe725354fe91091bfa10a667832ad964b24
- metric_population
- 295
- benchmark_evaluation
- true
- source model id
- Qwen2.5-VL-72B
Original result ↗97.5%
- source id
- mmBench
- leaderboard name
- celebrity_recognition
- dimension
- vision
- metric
- celebrity_recognition
- rank
- 2
- aggregation scope
- status
- published
- direction
- higherIsBetter
- conditions
- suite
- MMBench_TEST_CN_Data
- verified
- Yes
- metric_kind
- celebrity_recognition
- rank_method
- competition rank among reported evaluations in identical conditions, including unlinked observations
- source_file
- mmbench-data.json
- source_revision
- acb08c72ac038d24cf7785254678efe725354fe91091bfa10a667832ad964b24
- metric_population
- 163
- benchmark_evaluation
- true
- source model id
- Qwen2.5-VL-72B
Original result ↗97.5%
- source id
- mmBench
- leaderboard name
- celebrity_recognition
- dimension
- vision
- metric
- celebrity_recognition
- rank
- 2
- aggregation scope
- status
- published
- direction
- higherIsBetter
- conditions
- suite
- MMBench_TEST_EN_Data
- verified
- Yes
- metric_kind
- celebrity_recognition
- rank_method
- competition rank among reported evaluations in identical conditions, including unlinked observations
- source_file
- mmbench-data.json
- source_revision
- acb08c72ac038d24cf7785254678efe725354fe91091bfa10a667832ad964b24
- metric_population
- 162
- benchmark_evaluation
- true
- source model id
- Qwen2.5-VL-72B
Original result ↗92.4%
- source id
- mmBench
- leaderboard name
- object_localization
- dimension
- vision
- metric
- object_localization
- rank
- 2
- aggregation scope
- status
- published
- direction
- higherIsBetter
- conditions
- suite
- MMBench_TEST_CN_V11_Data
- verified
- Yes
- metric_kind
- object_localization
- rank_method
- competition rank among reported evaluations in identical conditions, including unlinked observations
- source_file
- mmbench-data.json
- source_revision
- acb08c72ac038d24cf7785254678efe725354fe91091bfa10a667832ad964b24
- metric_population
- 295
- benchmark_evaluation
- true
- source model id
- Qwen2.5-VL-72B
Original result ↗Published prices
1–3 / 3
$0.40
- Inference provider
- openRouter
- Platform
- OpenRouter
- Unit
- cache_read_tokens_per_million
- Currency
- USD
$0.80
- Inference provider
- openRouter
- Platform
- OpenRouter
- Unit
- input_tokens_per_million
- Currency
- USD
$1.00
- Inference provider
- openRouter
- Platform
- OpenRouter
- Unit
- output_tokens_per_million
- Currency
- USD
Provider specifications & access
1–3 / 3
- source model id
- HuggingFace/qwen-qwen2.5-vl-72b-instruct@1
- hosting
- foundry-managed-compute
- license
- apache-2.0
- source model id
- publishers/hf-qwen/models/qwen2.5-vl-72b-instruct@001
- hosting
- vertex-self-deployed
- source model id
- Qwen/Qwen2.5-VL-72B-Instruct
- hosting
- siliconflow-hosted
- max output tokens
- 4096
Safety assessments
1–1 / 1
26 / 100
- source id
- stanfordFMTI
- canonical model id
- qwen/qwen2.5-vl-72b-instruct
- taxonomy
- transparency
- hazard category
- transparency
- hazard name
- Organization transparency
- policy area
- Alibaba
- assessment type
- organization-level disclosure
- metric
- disclosure_score
- result label
- 26 / 100
- sample size
- 100
- updated at
- 2026-11-09T18:08:05.227Z
Source specification observations
1–5 / 5
- field
- context_length
- value
- 128000
- normalized value
- 128000
- scope
- source-model
- source id
- openRouter
- confidence
- 1
- field
- context_length
- value
- 131K
- scope
- source-model
- source id
- siliconFlow
- 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
- max_output_tokens
- value
- 115200
- normalized value
- 115200
- scope
- source-model
- source id
- openRouter
- confidence
- 1
- field
- max_output_tokens
- value
- 4K
- scope
- source-model
- source id
- siliconFlow
- confidence
- 1
Explore more models