# Code Generation on MBPP, MBPP+, BCB-f, BCB-h, and LCB Source: [Code Generation on MBPP, MBPP+, BCB-f, BCB-h, and LCB]() Collection index: [llms.txt]() Data updated: 2026-05-29T11:41:10.689Z Evaluation of model performance across a combined set of code generation benchmarks including MBPP, MBPP+, BCB-f, BCB-h, and LCB, reporting individual and average scores. 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 | Code Generation | | Dataset (including any stated split) | Code Generation Benchmarks MBPP, MBPP+, BCB-f, BCB-h, LCB | | Metric | MBPP Score | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 10 of 10 | | 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 - [MBPP Score](): Higher is better; unit: PERCENT. - [MBPP+ Score](): Higher is better; unit: PERCENT. - [BCB-f Score](): Higher is better; unit: PERCENT. - [BCB-h Score](): Higher is better; unit: PERCENT. - [LCB Score](): Higher is better; unit: PERCENT. - [Average Code Generation Score](): Higher is better; unit: PERCENT. ## Recorded results | Method | MBPP Score | Result date | Report paper | | --- | --- | --- | --- | | MIRA-Group | 88.9 | Not recorded | 6a191dbedbd7150506f82a17 | | MIRA-Global | 88.4 | Not recorded | 6a191dbedbd7150506f82a17 | | DataMan | 87.8 | Not recorded | 6a191dbedbd7150506f82a17 | | Raw Mixture | 87.6 | Not recorded | 6a191dbedbd7150506f82a17 | | MIRA-Source | 87.3 | Not recorded | 6a191dbedbd7150506f82a17 | | +SFT | 86.8 | Not recorded | 6a191dbedbd7150506f82a17 | | DSIR | 86.8 | Not recorded | 6a191dbedbd7150506f82a17 | | Random | 86.8 | Not recorded | 6a191dbedbd7150506f82a17 | | Base Model | 84.1 | Not recorded | 6a191dbedbd7150506f82a17 | | PPL | 83.6 | Not recorded | 6a191dbedbd7150506f82a17 | ### MIRA-Group MBPP Score: 88.9. Result ID: 6a197b4cfbc373592c14d3a5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Mid-training Protocol":"Subset Selection","Token Budget":"25B"}. This record does not assert independent reproduction. ### MIRA-Global MBPP Score: 88.4. Result ID: 6a197b4cfbc373592c14d3a4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Mid-training Protocol":"Subset Selection","Token Budget":"25B"}. This record does not assert independent reproduction. ### DataMan MBPP Score: 87.8. Result ID: 6a197b4cfbc373592c14d3a3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Mid-training Protocol":"Subset Selection","Token Budget":"25B"}. This record does not assert independent reproduction. ### Raw Mixture MBPP Score: 87.6. Result ID: 6a197b4cfbc373592c14d39f. [Reporting paper]() Recorded evaluation configuration: {"Mid-training Protocol":"Full Corpus","Token Budget":"50B"}. This record does not assert independent reproduction. ### MIRA-Source MBPP Score: 87.3. Result ID: 6a197b4cfbc373592c14d3a6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Mid-training Protocol":"Subset Selection","Token Budget":"25B"}. This record does not assert independent reproduction. ### +SFT MBPP Score: 86.8. Result ID: 6a197b4cfbc373592c14d39e. [Reporting paper]() Recorded evaluation configuration: {"Mid-training Protocol":"No Mid-training","Token Budget":"0"}. This record does not assert independent reproduction. ### DSIR MBPP Score: 86.8. Result ID: 6a197b4cfbc373592c14d3a0. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Mid-training Protocol":"Subset Selection","Token Budget":"25B"}. This record does not assert independent reproduction. ### Random MBPP Score: 86.8. Result ID: 6a197b4cfbc373592c14d3a2. [Reporting paper]() Recorded evaluation configuration: {"Mid-training Protocol":"Subset Selection","Token Budget":"25B"}. This record does not assert independent reproduction. ### Base Model MBPP Score: 84.1. Result ID: 6a197b4cfbc373592c14d39d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Mid-training Protocol":"No Mid-training","Token Budget":"0"}. This record does not assert independent reproduction. ### PPL MBPP Score: 83.6. Result ID: 6a197b4cfbc373592c14d3a1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Mid-training Protocol":"Subset Selection","Token Budget":"25B"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.