# Information Seeking Question Answering on InfoSeek Source: [Information Seeking Question Answering on InfoSeek]() Collection index: [llms.txt]() Data updated: 2026-03-04T13:04:01.419Z Evaluation of model performance on the InfoSeek dataset for Information Seeking Question Answering tasks. A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Information Seeking Question Answering | | Dataset (including any stated split) | InfoSeek | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 17 of 17 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | MM-DeepResearch 32B | 73.9 | Not recorded | 69a66c122d603e6de3e3cd77 | | MM-DeepResearch-8B | 72.3 | Not recorded | 69a66c122d603e6de3e3cd77 | | GPT-5 | 70.6 | Not recorded | 69a66c122d603e6de3e3cd77 | | SenseNova-MARS-8B | 70.1 | Not recorded | 69a66c122d603e6de3e3cd77 | | GPT-5 | 61.7 | Not recorded | 69a66c122d603e6de3e3cd77 | | GPT-4o | 59.5 | Not recorded | 69a66c122d603e6de3e3cd77 | | MM-DeepResearch-7B | 58.7 | Not recorded | 69a66c122d603e6de3e3cd77 | | Qwen3-VL-32B | 58.5 | Not recorded | 69a66c122d603e6de3e3cd77 | | MMSearch-R1-7B | 55.1 | Not recorded | 69a66c122d603e6de3e3cd77 | | GPT-4o | 52.9 | Not recorded | 69a66c122d603e6de3e3cd77 | | DeepEyes-v2-7B | 51.1 | Not recorded | 69a66c122d603e6de3e3cd77 | | Qwen3-VL-8B | 50.3 | Not recorded | 69a66c122d603e6de3e3cd77 | | DeepMMSearch-R1-7B | 47.5 | Not recorded | 69a66c122d603e6de3e3cd77 | | Qwen3-VL-8B | 46.1 | Not recorded | 69a66c122d603e6de3e3cd77 | | Visual-ARFT-7B | 37.9 | Not recorded | 69a66c122d603e6de3e3cd77 | | Qwen3-VL-32B | 28 | Not recorded | 69a66c122d603e6de3e3cd77 | | Qwen3-VL-8B | 23.1 | Not recorded | 69a66c122d603e6de3e3cd77 | ### MM-DeepResearch 32B Accuracy: 73.9. Result ID: 69a82daa6dbb7075b94e870f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Paradigm":"Agentic Search"}. This record does not assert independent reproduction. ### MM-DeepResearch-8B Accuracy: 72.3. Result ID: 69a82daa6dbb7075b94e870d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Paradigm":"Agentic Search"}. This record does not assert independent reproduction. ### GPT-5 Accuracy: 70.6. Result ID: 69a82daa6dbb7075b94e8704. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Paradigm":"RAG Workflow"}. This record does not assert independent reproduction. ### SenseNova-MARS-8B Accuracy: 70.1. Result ID: 69a82daa6dbb7075b94e870c. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Evaluation Paradigm":"Agentic Search"}. This record does not assert independent reproduction. ### GPT-5 Accuracy: 61.7. Result ID: 69a82daa6dbb7075b94e8700. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Paradigm":"Direct Reasoning"}. This record does not assert independent reproduction. ### GPT-4o Accuracy: 59.5. Result ID: 69a82daa6dbb7075b94e8703. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Paradigm":"RAG Workflow"}. This record does not assert independent reproduction. ### MM-DeepResearch-7B Accuracy: 58.7. Result ID: 69a82daa6dbb7075b94e870a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Paradigm":"Agentic Search"}. This record does not assert independent reproduction. ### Qwen3-VL-32B Accuracy: 58.5. Result ID: 69a82daa6dbb7075b94e870e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Paradigm":"Agentic Search"}. This record does not assert independent reproduction. ### MMSearch-R1-7B Accuracy: 55.1. Result ID: 69a82daa6dbb7075b94e8708. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Paradigm":"Agentic Search"}. This record does not assert independent reproduction. ### GPT-4o Accuracy: 52.9. Result ID: 69a82daa6dbb7075b94e86ff. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Paradigm":"Direct Reasoning"}. This record does not assert independent reproduction. ### DeepEyes-v2-7B Accuracy: 51.1. Result ID: 69a82daa6dbb7075b94e8709. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Evaluation Paradigm":"Agentic Search"}. This record does not assert independent reproduction. ### Qwen3-VL-8B Accuracy: 50.3. Result ID: 69a82daa6dbb7075b94e870b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Paradigm":"Agentic Search"}. This record does not assert independent reproduction. ### DeepMMSearch-R1-7B Accuracy: 47.5. Result ID: 69a82daa6dbb7075b94e8706. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Paradigm":"Agentic Search"}. This record does not assert independent reproduction. ### Qwen3-VL-8B Accuracy: 46.1. Result ID: 69a82daa6dbb7075b94e8705. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Paradigm":"RAG Workflow"}. This record does not assert independent reproduction. ### Visual-ARFT-7B Accuracy: 37.9. Result ID: 69a82daa6dbb7075b94e8707. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Paradigm":"Agentic Search"}. This record does not assert independent reproduction. ### Qwen3-VL-32B Accuracy: 28. Result ID: 69a82daa6dbb7075b94e8702. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Paradigm":"Direct Reasoning"}. This record does not assert independent reproduction. ### Qwen3-VL-8B Accuracy: 23.1. Result ID: 69a82daa6dbb7075b94e8701. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Paradigm":"Direct Reasoning"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.