# Multitask Language Understanding on MMMLU Swahili 1.0 (test) Source: [Multitask Language Understanding on MMMLU Swahili 1.0 (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:30:32.203Z Evaluation of model performance on the test split of the Swahili language version (1.0) of the MMLU benchmark for Multitask Language Understanding, reporting 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 | Multitask Language Understanding | | Dataset (including any stated split) | MMMLU Swahili 1.0 (test) | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 18 of 18 | | 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 | | --- | --- | --- | --- | | CLO | 33.38 | Not recorded | 68fa78ac1e209131cea79c93 | | CLO | 29.8 | Not recorded | 68fa78ac1e209131cea79c93 | | SFT-tgt | 29.05 | Not recorded | 68fa78ac1e209131cea79c93 | | SFT+DPO | 28.86 | Not recorded | 68fa78ac1e209131cea79c93 | | SFT-tgt | 28.61 | Not recorded | 68fa78ac1e209131cea79c93 | | SFT+DPO | 28.42 | Not recorded | 68fa78ac1e209131cea79c93 | | CLO | 28.07 | Not recorded | 68fa78ac1e209131cea79c93 | | SFT | 27.59 | Not recorded | 68fa78ac1e209131cea79c93 | | CLO | 26.78 | Not recorded | 68fa78ac1e209131cea79c93 | | SFT | 26.68 | Not recorded | 68fa78ac1e209131cea79c93 | | SFT | 26.22 | Not recorded | 68fa78ac1e209131cea79c93 | | CLO | 24.09 | Not recorded | 68fa78ac1e209131cea79c93 | | SFT-tgt | 22.56 | Not recorded | 68fa78ac1e209131cea79c93 | | SFT-tgt | 22.46 | Not recorded | 68fa78ac1e209131cea79c93 | | SFT | 21.38 | Not recorded | 68fa78ac1e209131cea79c93 | | SFT+DPO | 19.96 | Not recorded | 68fa78ac1e209131cea79c93 | | SFT+DPO | 19.83 | Not recorded | 68fa78ac1e209131cea79c93 | | SFT | 17.21 | Not recorded | 68fa78ac1e209131cea79c93 | ### CLO Accuracy: 33.38. Result ID: 69959b4a3badfaffe696f626. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Llama-3-8B","Training Data":"English + Swahili"}. This record does not assert independent reproduction. ### CLO Accuracy: 29.8. Result ID: 69959b4a3badfaffe696f62c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Qwen2.5-3B","Training Data":"English + Swahili"}. This record does not assert independent reproduction. ### SFT-tgt Accuracy: 29.05. Result ID: 69959b4a3badfaffe696f623. [Reporting paper]() Recorded evaluation configuration: {"Base Model":"Llama-3-8B","Training Data":"Target language only"}. This record does not assert independent reproduction. ### SFT+DPO Accuracy: 28.86. Result ID: 69959b4a3badfaffe696f625. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Llama-3-8B","Training Data":"English + Swahili"}. This record does not assert independent reproduction. ### SFT-tgt Accuracy: 28.61. Result ID: 69959b4a3badfaffe696f627. [Reporting paper]() Recorded evaluation configuration: {"Base Model":"Mistral-7B","Training Data":"Target language only"}. This record does not assert independent reproduction. ### SFT+DPO Accuracy: 28.42. Result ID: 69959b4a3badfaffe696f629. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Mistral-7B","Training Data":"English + Swahili"}. This record does not assert independent reproduction. ### CLO Accuracy: 28.07. Result ID: 69959b4a3badfaffe696f62a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Mistral-7B","Training Data":"English + Swahili"}. This record does not assert independent reproduction. ### SFT Accuracy: 27.59. Result ID: 69959b4a3badfaffe696f624. [Reporting paper]() Recorded evaluation configuration: {"Base Model":"Llama-3-8B","Training Data":"English + Swahili"}. This record does not assert independent reproduction. ### CLO Accuracy: 26.78. Result ID: 69959b4a3badfaffe696f622. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Llama-2-13B","Training Data":"English + Swahili"}. This record does not assert independent reproduction. ### SFT Accuracy: 26.68. Result ID: 69959b4a3badfaffe696f628. [Reporting paper]() Recorded evaluation configuration: {"Base Model":"Mistral-7B","Training Data":"English + Swahili"}. This record does not assert independent reproduction. ### SFT Accuracy: 26.22. Result ID: 69959b4a3badfaffe696f62b. [Reporting paper]() Recorded evaluation configuration: {"Base Model":"Qwen2.5-3B","Training Data":"English + Swahili"}. This record does not assert independent reproduction. ### CLO Accuracy: 24.09. Result ID: 69959b4a3badfaffe696f61e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Llama-2-7B","Training Data":"English + Swahili"}. This record does not assert independent reproduction. ### SFT-tgt Accuracy: 22.56. Result ID: 69959b4a3badfaffe696f61f. [Reporting paper]() Recorded evaluation configuration: {"Base Model":"Llama-2-13B","Training Data":"Target language only"}. This record does not assert independent reproduction. ### SFT-tgt Accuracy: 22.46. Result ID: 69959b4a3badfaffe696f61b. [Reporting paper]() Recorded evaluation configuration: {"Base Model":"Llama-2-7B","Training Data":"Target language only"}. This record does not assert independent reproduction. ### SFT Accuracy: 21.38. Result ID: 69959b4a3badfaffe696f620. [Reporting paper]() Recorded evaluation configuration: {"Base Model":"Llama-2-13B","Training Data":"English + Swahili"}. This record does not assert independent reproduction. ### SFT+DPO Accuracy: 19.96. Result ID: 69959b4a3badfaffe696f61d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Llama-2-7B","Training Data":"English + Swahili"}. This record does not assert independent reproduction. ### SFT+DPO Accuracy: 19.83. Result ID: 69959b4a3badfaffe696f621. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Llama-2-13B","Training Data":"English + Swahili"}. This record does not assert independent reproduction. ### SFT Accuracy: 17.21. Result ID: 69959b4a3badfaffe696f61c. [Reporting paper]() Recorded evaluation configuration: {"Base Model":"Llama-2-7B","Training Data":"English + Swahili"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.