{"work":{"id":"83f7c85b-3f11-450f-ac0c-64d9745220b2","openalex_id":"https://openalex.org/W1536441941","doi":"10.1038/nrn2888","arxiv_id":"2312.11805","raw_key":null,"title":"Gemini: A Family of Highly Capable Multimodal Models","authors":null,"authors_text":"Gemini Team, Rohan Anil, Sebastian Borgeaud, Jean-Baptiste Alayrac, Jiahui Yu, Radu Soricut, Johan Schalkwyk, Andrew M Dai, Anja Hauth, Katie Millican, et al","year":2023,"venue":"cs.CL","abstract":"This report introduces a new family of multimodal models, Gemini, that exhibit remarkable capabilities across image, audio, video, and text understanding. The Gemini family consists of Ultra, Pro, and Nano sizes, suitable for applications ranging from complex reasoning tasks to on-device memory-constrained use-cases. Evaluation on a broad range of benchmarks shows that our most-capable Gemini Ultra model advances the state of the art in 30 of 32 of these benchmarks - notably being the first model to achieve human-expert performance on the well-studied exam benchmark MMLU, and improving the state of the art in every one of the 20 multimodal benchmarks we examined. We believe that the new capabilities of the Gemini family in cross-modal reasoning and language understanding will enable a wide variety of use cases. We discuss our approach toward post-training and deploying Gemini models responsibly to users through services including Gemini, Gemini Advanced, Google AI Studio, and Cloud Vertex AI.","external_url":"https://arxiv.org/abs/2312.11805","cited_by_count":1,"metadata_source":"pith","metadata_fetched_at":"2026-08-05T02:28:24.338817+00:00","pith_arxiv_id":"2312.11805","created_at":"2026-05-09T03:55:07.747077+00:00","updated_at":"2026-08-05T02:28:24.338817+00:00","title_quality_ok":true,"display_title":"Gemini: A Family of Highly Capable Multimodal Models","render_title":"Gemini: A Family of Highly Capable Multimodal Models"},"hub":{"state":{"work_id":"83f7c85b-3f11-450f-ac0c-64d9745220b2","tier":"super_hub","tier_reason":"100+ Pith inbound or 10,000+ external citations","pith_inbound_count":964,"external_cited_by_count":1,"distinct_field_count":41,"first_pith_cited_at":"2023-07-12T16:23:09+00:00","last_pith_cited_at":"2026-07-09T13:55:55+00:00","author_build_status":"needed","summary_status":"needed","contexts_status":"needed","graph_status":"needed","ask_index_status":"needed","reader_status":"not_needed","recognition_status":"not_needed","updated_at":"2026-08-22T10:49:24.223145+00:00","tier_text":"super_hub"},"tier":"super_hub","role_counts":[{"context_role":"background","n":143},{"context_role":"baseline","n":33},{"context_role":"method","n":16},{"context_role":"dataset","n":4},{"context_role":"other","n":2}],"polarity_counts":[{"context_polarity":"background","n":138},{"context_polarity":"baseline","n":33},{"context_polarity":"use_method","n":16},{"context_polarity":"unclear","n":6},{"context_polarity":"use_dataset","n":4},{"context_polarity":"support","n":1}],"runs":{"ask_index":{"job_type":"ask_index","status":"succeeded","result":{"title":"Gemini: A Family of Highly Capable Multimodal Models","claims":[{"claim_text":"This report introduces a new family of multimodal models, Gemini, that exhibit remarkable capabilities across image, audio, video, and text understanding. The Gemini family consists of Ultra, Pro, and Nano sizes, suitable for applications ranging from complex reasoning tasks to on-device memory-constrained use-cases. Evaluation on a broad range of benchmarks shows that our most-capable Gemini Ultra model advances the state of the art in 30 of 32 of these benchmarks - notably being the first model to achieve human-expert performance on the well-studied exam benchmark MMLU, and improving the sta","claim_type":"abstract","evidence_strength":"source_metadata"}],"why_cited":"Pith tracks Gemini: A Family of Highly Capable Multimodal Models because it crossed a citation-hub threshold.","role_counts":[]},"error":null,"updated_at":"2026-05-13T18:53:25.894399+00:00"},"author_expand":{"job_type":"author_expand","status":"succeeded","result":{"authors_linked":[{"id":"68fcab02-5ea2-4d5a-bee1-04b66eb9b2e4","orcid":null,"display_name":"Gemini Team"},{"id":"90c56405-ec27-4260-ba84-1a6ca3790056","orcid":null,"display_name":"Rohan Anil"},{"id":"85fb8136-11ae-4eb2-9893-62a53fa26fb7","orcid":null,"display_name":"Sebastian Borgeaud"},{"id":"44b3ddc4-9896-4039-9349-0e6fe7adb78c","orcid":null,"display_name":"Jean-Baptiste Alayrac"},{"id":"87515ce0-f325-4b40-a0a8-6a2ad226b142","orcid":null,"display_name":"Jiahui Yu"},{"id":"cc9b94b7-c09d-404c-9b1f-e800e61a08df","orcid":null,"display_name":"Radu Soricut"}]},"error":null,"updated_at":"2026-05-13T18:53:25.890249+00:00"},"context_extract":{"job_type":"context_extract","status":"succeeded","result":{"enqueued_papers":25},"error":null,"updated_at":"2026-05-13T18:43:32.815940+00:00"},"graph_features":{"job_type":"graph_features","status":"succeeded","result":{"co_cited":[{"title":"GPT-4 Technical Report","work_id":"b928e041-6991-4c08-8c81-0359e4097c7b","shared_citers":116},{"title":"Qwen3 Technical Report","work_id":"25a4e30c-1232-48e7-9925-02fa12ba7c9e","shared_citers":51},{"title":"The Llama 3 Herd of Models","work_id":"1549a635-88af-4ac1-acfe-51ae7bb53345","shared_citers":49},{"title":"Qwen3-VL Technical Report","work_id":"1fe243aa-e3c0-4da6-b391-4cbcfc88d5c0","shared_citers":45},{"title":"LLaMA: Open and Efficient Foundation Language Models","work_id":"c018fc23-6f3f-4035-9d02-28a2173b2b9d","shared_citers":44},{"title":"GPT-4o System Card","work_id":"f37bf1c7-4964-4e56-9762-d20da8d9009f","shared_citers":43},{"title":"Llama 2: Open Foundation and Fine-Tuned Chat Models","work_id":"68a5177f-d644-44c1-bd4f-4e5278c22f5d","shared_citers":40},{"title":"Qwen2.5-VL Technical Report","work_id":"69dffacb-bfe8-442d-be86-48624c60426f","shared_citers":38},{"title":"Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution","work_id":"8abcfe4f-e0fb-44b7-9123-448fac95f90a","shared_citers":37},{"title":"DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning","work_id":"e6b75ad5-2877-4168-97c8-710407094d20","shared_citers":32},{"title":"Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities","work_id":"008df105-2fdd-45d8-857a-8e35868aecb6","shared_citers":32},{"title":"Qwen Technical Report","work_id":"bb1fd52f-6b2f-437c-9516-37bdf6eb9be8","shared_citers":30},{"title":"DeepSeek-V3 Technical Report","work_id":"57d2791d-2219-4c31-a077-afc04b12a75c","shared_citers":27},{"title":"Evaluating Large Language Models Trained on Code","work_id":"042493e9-b26f-4b4e-bbde-382072ca9b08","shared_citers":26},{"title":"Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyond","work_id":"cbc2bb21-b6bb-46c0-80bf-107e195ffe10","shared_citers":25},{"title":"Training Verifiers to Solve Math Word Problems","work_id":"acab1aa8-b4d6-40e0-a3ee-25341701dca2","shared_citers":24},{"title":"DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models","work_id":"c5006563-f3ec-438a-9e35-b7b484f34828","shared_citers":22},{"title":"Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context","work_id":"80e3e977-f1bb-4c83-8d0c-1ab0a0c5c3f1","shared_citers":21},{"title":"OpenAI GPT-5 System Card","work_id":"ca87689a-0d29-4476-b504-b65dbbb08af4","shared_citers":21},{"title":"Qwen2.5 Technical Report","work_id":"d8432992-4980-4a81-85c7-9fa2c2b87f85","shared_citers":21},{"title":"LLaVA-OneVision: Easy Visual Task Transfer","work_id":"f5f2452b-f2a9-49ac-b38d-c76e18cdfe49","shared_citers":20},{"title":"Scaling Laws for Neural Language Models","work_id":"b7dd8749-9c45-4977-ab9b-64478dce1ae8","shared_citers":20},{"title":"InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiency","work_id":"b8f5e260-fff5-444e-bcf5-2c42cfefd83d","shared_citers":18},{"title":"MathVista: Evaluating Mathematical Reasoning of Foundation Models in Visual Contexts","work_id":"e22c3789-9e71-4242-b6ea-3e60e06e2b66","shared_citers":18}],"time_series":[{"n":3,"year":2023},{"n":28,"year":2024},{"n":13,"year":2025},{"n":262,"year":2026}]},"error":null,"updated_at":"2026-05-13T18:53:25.479398+00:00"},"identity_refresh":{"job_type":"identity_refresh","status":"succeeded","result":{"fixed":1,"items":[{"title":"Qwen3 Technical Report","work_id":"25a4e30c-1232-48e7-9925-02fa12ba7c9e","resolver":"local_arxiv","confidence":0.98,"old_work_id":"25a4e30c-1232-48e7-9925-02fa12ba7c9e"}],"errors":[],"attempted":1},"error":null,"updated_at":"2026-05-13T18:43:32.199475+00:00"},"role_polarity":{"job_type":"role_polarity","status":"succeeded","result":{"title":"Gemini: A Family of Highly Capable Multimodal Models","claims":[{"claim_text":"This report introduces a new family of multimodal models, Gemini, that exhibit remarkable capabilities across image, audio, video, and text understanding. The Gemini family consists of Ultra, Pro, and Nano sizes, suitable for applications ranging from complex reasoning tasks to on-device memory-constrained use-cases. Evaluation on a broad range of benchmarks shows that our most-capable Gemini Ultra model advances the state of the art in 30 of 32 of these benchmarks - notably being the first model to achieve human-expert performance on the well-studied exam benchmark MMLU, and improving the sta","claim_type":"abstract","evidence_strength":"source_metadata"}],"why_cited":"Pith tracks Gemini: A Family of Highly Capable Multimodal Models because it crossed a citation-hub threshold.","role_counts":[]},"error":null,"updated_at":"2026-05-13T18:43:32.818099+00:00"},"summary_claims":{"job_type":"summary_claims","status":"succeeded","result":{"title":"Gemini: A Family of Highly Capable Multimodal Models","claims":[{"claim_text":"This report introduces a new family of multimodal models, Gemini, that exhibit remarkable capabilities across image, audio, video, and text understanding. The Gemini family consists of Ultra, Pro, and Nano sizes, suitable for applications ranging from complex reasoning tasks to on-device memory-constrained use-cases. Evaluation on a broad range of benchmarks shows that our most-capable Gemini Ultra model advances the state of the art in 30 of 32 of these benchmarks - notably being the first model to achieve human-expert performance on the well-studied exam benchmark MMLU, and improving the sta","claim_type":"abstract","evidence_strength":"source_metadata"}],"why_cited":"Pith tracks Gemini: A Family of Highly Capable Multimodal Models because it crossed a citation-hub threshold.","role_counts":[]},"error":null,"updated_at":"2026-05-13T17:25:52.730580+00:00"}},"summary":{"title":"Gemini: A Family of Highly Capable Multimodal Models","claims":[{"claim_text":"This report introduces a new family of multimodal models, Gemini, that exhibit remarkable capabilities across image, audio, video, and text understanding. The Gemini family consists of Ultra, Pro, and Nano sizes, suitable for applications ranging from complex reasoning tasks to on-device memory-constrained use-cases. Evaluation on a broad range of benchmarks shows that our most-capable Gemini Ultra model advances the state of the art in 30 of 32 of these benchmarks - notably being the first model to achieve human-expert performance on the well-studied exam benchmark MMLU, and improving the sta","claim_type":"abstract","evidence_strength":"source_metadata"}],"why_cited":"Pith tracks Gemini: A Family of Highly Capable Multimodal Models because it crossed a citation-hub threshold.","role_counts":[]},"graph":{"co_cited":[{"title":"GPT-4 Technical Report","work_id":"b928e041-6991-4c08-8c81-0359e4097c7b","shared_citers":116},{"title":"Qwen3 Technical Report","work_id":"25a4e30c-1232-48e7-9925-02fa12ba7c9e","shared_citers":51},{"title":"The Llama 3 Herd of Models","work_id":"1549a635-88af-4ac1-acfe-51ae7bb53345","shared_citers":49},{"title":"Qwen3-VL Technical Report","work_id":"1fe243aa-e3c0-4da6-b391-4cbcfc88d5c0","shared_citers":45},{"title":"LLaMA: Open and Efficient Foundation Language Models","work_id":"c018fc23-6f3f-4035-9d02-28a2173b2b9d","shared_citers":44},{"title":"GPT-4o System Card","work_id":"f37bf1c7-4964-4e56-9762-d20da8d9009f","shared_citers":43},{"title":"Llama 2: Open Foundation and Fine-Tuned Chat Models","work_id":"68a5177f-d644-44c1-bd4f-4e5278c22f5d","shared_citers":40},{"title":"Qwen2.5-VL Technical Report","work_id":"69dffacb-bfe8-442d-be86-48624c60426f","shared_citers":38},{"title":"Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution","work_id":"8abcfe4f-e0fb-44b7-9123-448fac95f90a","shared_citers":37},{"title":"DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning","work_id":"e6b75ad5-2877-4168-97c8-710407094d20","shared_citers":32},{"title":"Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities","work_id":"008df105-2fdd-45d8-857a-8e35868aecb6","shared_citers":32},{"title":"Qwen Technical Report","work_id":"bb1fd52f-6b2f-437c-9516-37bdf6eb9be8","shared_citers":30},{"title":"DeepSeek-V3 Technical Report","work_id":"57d2791d-2219-4c31-a077-afc04b12a75c","shared_citers":27},{"title":"Evaluating Large Language Models Trained on Code","work_id":"042493e9-b26f-4b4e-bbde-382072ca9b08","shared_citers":26},{"title":"Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyond","work_id":"cbc2bb21-b6bb-46c0-80bf-107e195ffe10","shared_citers":25},{"title":"Training Verifiers to Solve Math Word Problems","work_id":"acab1aa8-b4d6-40e0-a3ee-25341701dca2","shared_citers":24},{"title":"DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models","work_id":"c5006563-f3ec-438a-9e35-b7b484f34828","shared_citers":22},{"title":"Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context","work_id":"80e3e977-f1bb-4c83-8d0c-1ab0a0c5c3f1","shared_citers":21},{"title":"OpenAI GPT-5 System Card","work_id":"ca87689a-0d29-4476-b504-b65dbbb08af4","shared_citers":21},{"title":"Qwen2.5 Technical Report","work_id":"d8432992-4980-4a81-85c7-9fa2c2b87f85","shared_citers":21},{"title":"LLaVA-OneVision: Easy Visual Task Transfer","work_id":"f5f2452b-f2a9-49ac-b38d-c76e18cdfe49","shared_citers":20},{"title":"Scaling Laws for Neural Language Models","work_id":"b7dd8749-9c45-4977-ab9b-64478dce1ae8","shared_citers":20},{"title":"InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiency","work_id":"b8f5e260-fff5-444e-bcf5-2c42cfefd83d","shared_citers":18},{"title":"MathVista: Evaluating Mathematical Reasoning of Foundation Models in Visual Contexts","work_id":"e22c3789-9e71-4242-b6ea-3e60e06e2b66","shared_citers":18}],"time_series":[{"n":3,"year":2023},{"n":28,"year":2024},{"n":13,"year":2025},{"n":262,"year":2026}]},"authors":[{"id":"68fcab02-5ea2-4d5a-bee1-04b66eb9b2e4","orcid":null,"display_name":"Gemini Team","source":"manual","import_confidence":0.72},{"id":"44b3ddc4-9896-4039-9349-0e6fe7adb78c","orcid":null,"display_name":"Jean-Baptiste Alayrac","source":"manual","import_confidence":0.72},{"id":"87515ce0-f325-4b40-a0a8-6a2ad226b142","orcid":null,"display_name":"Jiahui Yu","source":"manual","import_confidence":0.72},{"id":"cc9b94b7-c09d-404c-9b1f-e800e61a08df","orcid":null,"display_name":"Radu Soricut","source":"manual","import_confidence":0.72},{"id":"90c56405-ec27-4260-ba84-1a6ca3790056","orcid":null,"display_name":"Rohan Anil","source":"manual","import_confidence":0.72},{"id":"85fb8136-11ae-4eb2-9893-62a53fa26fb7","orcid":null,"display_name":"Sebastian Borgeaud","source":"manual","import_confidence":0.72}]}}