# Fill-in-the-Middle Code Completion on Delulu Source: [Fill-in-the-Middle Code Completion on Delulu]() Collection index: [llms.txt]() Data updated: 2026-06-04T08:08:33.272Z Evaluation of model performance on the Delulu dataset for Fill-in-the-Middle Code Completion tasks, measured using Exact Match (EM) and Edit Similarity (ES). 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 | Fill-in-the-Middle Code Completion | | Dataset (including any stated split) | Delulu | | Metric | EM (%) | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 16 of 16 | | 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 - [EM (%)](): Higher is better; unit: PERCENT. - [Edit Similarity](): Higher is better; unit: PERCENT. ## Recorded results | Method | EM (%) | Result date | Report paper | | --- | --- | --- | --- | | ORPO | 62.4 | Not recorded | 6a21065bc10484bac23efe7f | | ORPO | 62.4 | Not recorded | 6a21065bc10484bac23efe7f | | SFT | 61.8 | Not recorded | 6a21065bc10484bac23efe7f | | SFT | 61.7 | Not recorded | 6a21065bc10484bac23efe7f | | ORPO | 59.9 | Not recorded | 6a21065bc10484bac23efe7f | | ORPO | 59.7 | Not recorded | 6a21065bc10484bac23efe7f | | SFT | 59.2 | Not recorded | 6a21065bc10484bac23efe7f | | SFT | 58.7 | Not recorded | 6a21065bc10484bac23efe7f | | base | 52 | Not recorded | 6a21065bc10484bac23efe7f | | base | 48.8 | Not recorded | 6a21065bc10484bac23efe7f | | base | 45.9 | Not recorded | 6a21065bc10484bac23efe7f | | base | 42.9 | Not recorded | 6a21065bc10484bac23efe7f | | DPO | 25.8 | Not recorded | 6a21065bc10484bac23efe7f | | DPO | 24.2 | Not recorded | 6a21065bc10484bac23efe7f | | DPO | 16.9 | Not recorded | 6a21065bc10484bac23efe7f | | DPO | 13.3 | Not recorded | 6a21065bc10484bac23efe7f | ### ORPO EM (%): 62.4. Result ID: 6a2131a7fbc373592c1c9933. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Size":"7B","Truncation Strategy":"untruncated"}. This record does not assert independent reproduction. ### ORPO EM (%): 62.4. Result ID: 6a2131a7fbc373592c1c9934. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Size":"7B","Truncation Strategy":"+trunc"}. This record does not assert independent reproduction. ### SFT EM (%): 61.8. Result ID: 6a2131a7fbc373592c1c9930. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Size":"7B","Truncation Strategy":"+trunc"}. This record does not assert independent reproduction. ### SFT EM (%): 61.7. Result ID: 6a2131a7fbc373592c1c992f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Size":"7B","Truncation Strategy":"untruncated"}. This record does not assert independent reproduction. ### ORPO EM (%): 59.9. Result ID: 6a2131a7fbc373592c1c992c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Size":"3B","Truncation Strategy":"+trunc"}. This record does not assert independent reproduction. ### ORPO EM (%): 59.7. Result ID: 6a2131a7fbc373592c1c992b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Size":"3B","Truncation Strategy":"untruncated"}. This record does not assert independent reproduction. ### SFT EM (%): 59.2. Result ID: 6a2131a7fbc373592c1c9928. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Size":"3B","Truncation Strategy":"+trunc"}. This record does not assert independent reproduction. ### SFT EM (%): 58.7. Result ID: 6a2131a7fbc373592c1c9927. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Size":"3B","Truncation Strategy":"untruncated"}. This record does not assert independent reproduction. ### base EM (%): 52. Result ID: 6a2131a7fbc373592c1c9926. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Size":"3B","Truncation Strategy":"+trunc"}. This record does not assert independent reproduction. ### base EM (%): 48.8. Result ID: 6a2131a7fbc373592c1c992e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Size":"7B","Truncation Strategy":"+trunc"}. This record does not assert independent reproduction. ### base EM (%): 45.9. Result ID: 6a2131a7fbc373592c1c9925. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Size":"3B","Truncation Strategy":"untruncated"}. This record does not assert independent reproduction. ### base EM (%): 42.9. Result ID: 6a2131a7fbc373592c1c992d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Size":"7B","Truncation Strategy":"untruncated"}. This record does not assert independent reproduction. ### DPO EM (%): 25.8. Result ID: 6a2131a7fbc373592c1c9932. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Size":"7B","Truncation Strategy":"+trunc"}. This record does not assert independent reproduction. ### DPO EM (%): 24.2. Result ID: 6a2131a7fbc373592c1c9931. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Size":"7B","Truncation Strategy":"untruncated"}. This record does not assert independent reproduction. ### DPO EM (%): 16.9. Result ID: 6a2131a7fbc373592c1c992a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Size":"3B","Truncation Strategy":"+trunc"}. This record does not assert independent reproduction. ### DPO EM (%): 13.3. Result ID: 6a2131a7fbc373592c1c9929. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Size":"3B","Truncation Strategy":"untruncated"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.