# General Tasks on MMLU-redux Source: [General Tasks on MMLU-redux]() Collection index: [llms.txt]() Data updated: 2026-05-20T07:24:10.332Z Evaluation of model performance on the General Tasks category within the MMLU Redux dataset, measured by accuracy. 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 | General Tasks | | Dataset (including any stated split) | MMLU-redux | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 15 of 15 | | 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 | | --- | --- | --- | --- | | Qwen2.5-72B | 86.8 | Not recorded | 68fa773a1e209131cea71d3c | | Llama-3.1-405B | 86.2 | Not recorded | 68fa773a1e209131cea71d3c | | Qwen2.5-VL-72B | 85.9 | Not recorded | 68fa773a1e209131cea71d3c | | Llama-3.1-70B | 83 | Not recorded | 68fa773a1e209131cea71d3c | | Qwen2-72B | 81.6 | Not recorded | 68fa773a1e209131cea71d3c | | Fine-tuned | 76.6 | Not recorded | 6a0d4017f984739ed6737810 | | DiDi-Merg.-L | 76.2 | Not recorded | 6a0d4017f984739ed6737810 | | FREE-Merging | 75.2 | Not recorded | 6a0d4017f984739ed6737810 | | Zero-shot | 75.1 | Not recorded | 6a0d4017f984739ed6737810 | | Twin-Merging | 74.8 | Not recorded | 6a0d4017f984739ed6737810 | | Fine-tuned | 73 | Not recorded | 6a0d4017f984739ed6737810 | | DiDi-Merg.-L | 72.8 | Not recorded | 6a0d4017f984739ed6737810 | | Twin-Merging | 72.6 | Not recorded | 6a0d4017f984739ed6737810 | | FREE-Merging | 72.4 | Not recorded | 6a0d4017f984739ed6737810 | | Zero-shot | 70.5 | Not recorded | 6a0d4017f984739ed6737810 | ### Qwen2.5-72B Accuracy: 86.8. Result ID: 69947b3b3a6c3148491d838f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Type":"Instruct","Parameters":"72B"}. This record does not assert independent reproduction. ### Llama-3.1-405B Accuracy: 86.2. Result ID: 69947b3b3a6c3148491d838d. [Reporting paper]() Recorded evaluation configuration: {"Model Type":"Instruct","Parameters":"405B"}. This record does not assert independent reproduction. ### Qwen2.5-VL-72B Accuracy: 85.9. Result ID: 69947b3b3a6c3148491d8390. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Type":"Instruct","Parameters":"72B"}. This record does not assert independent reproduction. ### Llama-3.1-70B Accuracy: 83. Result ID: 69947b3b3a6c3148491d838c. [Reporting paper]() Recorded evaluation configuration: {"Model Type":"Instruct","Parameters":"70B"}. This record does not assert independent reproduction. ### Qwen2-72B Accuracy: 81.6. Result ID: 69947b3b3a6c3148491d838e. [Reporting paper]() Recorded evaluation configuration: {"Model Type":"Instruct","Parameters":"72B"}. This record does not assert independent reproduction. ### Fine-tuned Accuracy: 76.6. Result ID: 6a0d6180fbc373592c050740. [Reporting paper]() Recorded evaluation configuration: {"Base LLM":"Qwen-2.5-7B-Instruct","Parameters":"x4"}. This record does not assert independent reproduction. ### DiDi-Merg.-L Accuracy: 76.2. Result ID: 6a0d6180fbc373592c050743. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base LLM":"Qwen-2.5-7B-Instruct","Parameters":"x2.0"}. This record does not assert independent reproduction. ### FREE-Merging Accuracy: 75.2. Result ID: 6a0d6180fbc373592c050742. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base LLM":"Qwen-2.5-7B-Instruct","Parameters":"x2.08"}. This record does not assert independent reproduction. ### Zero-shot Accuracy: 75.1. Result ID: 6a0d6180fbc373592c05073f. [Reporting paper]() Recorded evaluation configuration: {"Base LLM":"Qwen-2.5-7B-Instruct","Parameters":"x1"}. This record does not assert independent reproduction. ### Twin-Merging Accuracy: 74.8. Result ID: 6a0d6180fbc373592c050741. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base LLM":"Qwen-2.5-7B-Instruct","Parameters":"x2.25"}. This record does not assert independent reproduction. ### Fine-tuned Accuracy: 73. Result ID: 6a0d6180fbc373592c05073b. [Reporting paper]() Recorded evaluation configuration: {"Base LLM":"Llama-3.1-8B-Instruct","Parameters":"x4"}. This record does not assert independent reproduction. ### DiDi-Merg.-L Accuracy: 72.8. Result ID: 6a0d6180fbc373592c05073e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base LLM":"Llama-3.1-8B-Instruct","Parameters":"x2.0"}. This record does not assert independent reproduction. ### Twin-Merging Accuracy: 72.6. Result ID: 6a0d6180fbc373592c05073c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base LLM":"Llama-3.1-8B-Instruct","Parameters":"x2.25"}. This record does not assert independent reproduction. ### FREE-Merging Accuracy: 72.4. Result ID: 6a0d6180fbc373592c05073d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base LLM":"Llama-3.1-8B-Instruct","Parameters":"x2.08"}. This record does not assert independent reproduction. ### Zero-shot Accuracy: 70.5. Result ID: 6a0d6180fbc373592c05073a. [Reporting paper]() Recorded evaluation configuration: {"Base LLM":"Llama-3.1-8B-Instruct","Parameters":"x1"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.