# Logic Reasoning on K&K Logic Puzzles In-domain Source: [Logic Reasoning on K&K Logic Puzzles In-domain]() Collection index: [llms.txt]() Data updated: 2026-03-10T09:32:35.278Z Evaluation of model performance on the in-domain logic puzzles from the K&K Logic Puzzles dataset, focusing on reasoning capabilities across different difficulty levels (3 through 7). 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 | Logic Reasoning | | Dataset (including any stated split) | K&K Logic Puzzles In-domain | | Metric | Accuracy (Level 3) | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 25 of 25 | | 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 (Level 3)](): Higher is better; unit: PERCENT. - [Accuracy (Level 4)](): Higher is better; unit: PERCENT. - [Accuracy (Level 5)](): Higher is better; unit: PERCENT. - [Accuracy (Level 6)](): Higher is better; unit: PERCENT. - [Accuracy (Level 7)](): Higher is better; unit: PERCENT. - [Average Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy (Level 3) | Result date | Report paper | | --- | --- | --- | --- | | o3-mini-high | 98 | Not recorded | 68fa7b961e209131cea894b1 | | MeRF | 95 | Not recorded | 68fa7b961e209131cea894b1 | | MeRF | 95 | Not recorded | 68fa7b961e209131cea894b1 | | RLVR | 92 | Not recorded | 68fa7b961e209131cea894b1 | | RLVR | 90 | Not recorded | 68fa7b961e209131cea894b1 | | MeRF | 78 | Not recorded | 68fa7b961e209131cea894b1 | | Deepseek-R1 | 73 | Not recorded | 68fa7b961e209131cea894b1 | | RLVR | 68 | Not recorded | 68fa7b961e209131cea894b1 | | MeRF | 68 | Not recorded | 68fa7b961e209131cea894b1 | | GPT-4o | 57 | Not recorded | 68fa7b961e209131cea894b1 | | MeRF | 54 | Not recorded | 68fa7b961e209131cea894b1 | | RLVR | 52 | Not recorded | 68fa7b961e209131cea894b1 | | o1-2024-12-17 | 51 | Not recorded | 68fa7b961e209131cea894b1 | | RLVR | 48 | Not recorded | 68fa7b961e209131cea894b1 | | Qwen2.5-14B-Instruct | 46 | Not recorded | 68fa7b961e209131cea894b1 | | GPT-4o-mini | 42 | Not recorded | 68fa7b961e209131cea894b1 | | Qwen2.5-7B-Instruct-1M | 40 | Not recorded | 68fa7b961e209131cea894b1 | | Qwen2.5-7B-Base | 34 | Not recorded | 68fa7b961e209131cea894b1 | | DeepSeek-R1-Distill-Llama-8B | 26 | Not recorded | 68fa7b961e209131cea894b1 | | MeRF | 25 | Not recorded | 68fa7b961e209131cea894b1 | | Qwen2.5-7B-Instruct | 24 | Not recorded | 68fa7b961e209131cea894b1 | | Deepseek-Math-7B | 21 | Not recorded | 68fa7b961e209131cea894b1 | | RLVR | 20 | Not recorded | 68fa7b961e209131cea894b1 | | NuminaMath-7B-CoT | 13 | Not recorded | 68fa7b961e209131cea894b1 | | DeepSeek-R1-Distill-Qwen-1.5B | 8 | Not recorded | 68fa7b961e209131cea894b1 | ### o3-mini-high Accuracy (Level 3): 98. Result ID: 69afe52b6dbb7075b9589dfb. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### MeRF Accuracy (Level 3): 95. Result ID: 69afe52b6dbb7075b9589e0d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-14B-Instruct"}. This record does not assert independent reproduction. ### MeRF Accuracy (Level 3): 95. Result ID: 69afe52b6dbb7075b9589e13. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Llama-8B"}. This record does not assert independent reproduction. ### RLVR Accuracy (Level 3): 92. Result ID: 69afe52b6dbb7075b9589e12. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Llama-8B"}. This record does not assert independent reproduction. ### RLVR Accuracy (Level 3): 90. Result ID: 69afe52b6dbb7075b9589e0c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-14B-Instruct"}. This record does not assert independent reproduction. ### MeRF Accuracy (Level 3): 78. Result ID: 69afe52b6dbb7075b9589e07. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B-Instruct"}. This record does not assert independent reproduction. ### Deepseek-R1 Accuracy (Level 3): 73. Result ID: 69afe52b6dbb7075b9589dfd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### RLVR Accuracy (Level 3): 68. Result ID: 69afe52b6dbb7075b9589e06. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B-Instruct"}. This record does not assert independent reproduction. ### MeRF Accuracy (Level 3): 68. Result ID: 69afe52b6dbb7075b9589e0a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B-Instruct-1M"}. This record does not assert independent reproduction. ### GPT-4o Accuracy (Level 3): 57. Result ID: 69afe52b6dbb7075b9589dfe. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### MeRF Accuracy (Level 3): 54. Result ID: 69afe52b6dbb7075b9589e04. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B-Base"}. This record does not assert independent reproduction. ### RLVR Accuracy (Level 3): 52. Result ID: 69afe52b6dbb7075b9589e09. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B-Instruct-1M"}. This record does not assert independent reproduction. ### o1-2024-12-17 Accuracy (Level 3): 51. Result ID: 69afe52b6dbb7075b9589dfc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### RLVR Accuracy (Level 3): 48. Result ID: 69afe52b6dbb7075b9589e03. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B-Base"}. This record does not assert independent reproduction. ### Qwen2.5-14B-Instruct Accuracy (Level 3): 46. Result ID: 69afe52b6dbb7075b9589e0b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### GPT-4o-mini Accuracy (Level 3): 42. Result ID: 69afe52b6dbb7075b9589dff. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Qwen2.5-7B-Instruct-1M Accuracy (Level 3): 40. Result ID: 69afe52b6dbb7075b9589e08. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Qwen2.5-7B-Base Accuracy (Level 3): 34. Result ID: 69afe52b6dbb7075b9589e02. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### DeepSeek-R1-Distill-Llama-8B Accuracy (Level 3): 26. Result ID: 69afe52b6dbb7075b9589e11. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### MeRF Accuracy (Level 3): 25. Result ID: 69afe52b6dbb7075b9589e10. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Qwen-1.5B"}. This record does not assert independent reproduction. ### Qwen2.5-7B-Instruct Accuracy (Level 3): 24. Result ID: 69afe52b6dbb7075b9589e05. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Deepseek-Math-7B Accuracy (Level 3): 21. Result ID: 69afe52b6dbb7075b9589e01. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### RLVR Accuracy (Level 3): 20. Result ID: 69afe52b6dbb7075b9589e0f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Qwen-1.5B"}. This record does not assert independent reproduction. ### NuminaMath-7B-CoT Accuracy (Level 3): 13. Result ID: 69afe52b6dbb7075b9589e00. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### DeepSeek-R1-Distill-Qwen-1.5B Accuracy (Level 3): 8. Result ID: 69afe52b6dbb7075b9589e0e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.