evalstate/diffusers-pr

1{2  "schema_version": "1.0",3  "repo": "huggingface/diffusers",4  "snapshot_id": "20260427T113512Z",5  "generated_at": "2026-04-27T12:02:03Z",6  "model": "gpt-5.4-mini?service_tier=flex",7  "reviews": [8    {9      "label": "issue",10      "component_index": 1,11      "component_count": 102,12      "review_unit_index": 1,13      "review_unit_count": 364,14      "status": "reviewed",15      "reason": null,16      "source": "cache",17      "cache_hit": true,18      "model": "gpt-5.4-mini?service_tier=flex",19      "cluster_id": "cluster-50-3014",20      "nodes": [21        "issue:10035",22        "issue:10091",23        "issue:10343",24        "issue:10804",25        "issue:10954",26        "issue:11083",27        "issue:11169",28        "issue:11225",29        "issue:11417",30        "issue:12044",31        "issue:12436",32        "issue:12480",33        "issue:12481",34        "issue:2044",35        "issue:2780",36        "issue:2819",37        "issue:2995",38        "issue:3809",39        "issue:3954",40        "issue:4104",41        "issue:4325",42        "issue:4557",43        "issue:4649",44        "issue:4754",45        "issue:4868",46        "issue:5082",47        "issue:5259",48        "issue:5295",49        "issue:5475",50        "issue:5510",51        "issue:5805",52        "issue:6324",53        "issue:6363",54        "issue:6517",55        "issue:6594",56        "issue:6727",57        "issue:6812",58        "issue:6815",59        "issue:6996",60        "issue:7330",61        "issue:7854",62        "issue:8236",63        "issue:8759",64        "issue:9293",65        "issue:9298",66        "issue:9546",67        "issue:9726",68        "issue:9727"69      ],70      "soft_pairs": [71        "issue:12480|issue:12481",72        "issue:5295|issue:5475",73        "issue:10035|issue:10091",74        "issue:5510|issue:7854",75        "issue:3954|issue:8236",76        "issue:12436|issue:12480",77        "issue:12436|issue:12481",78        "issue:6815|issue:8759",79        "issue:5805|issue:7854",80        "issue:10804|issue:10954",81        "issue:4325|issue:8759",82        "issue:4754|issue:5259",83        "issue:6727|issue:6812",84        "issue:4649|issue:6996",85        "issue:11083|issue:3809",86        "issue:12044|issue:6324",87        "issue:10343|issue:6815",88        "issue:2995|issue:5082",89        "issue:2819|issue:6594",90        "issue:9293|issue:9298",91        "issue:4557|issue:4868",92        "issue:11225|issue:9546",93        "issue:9726|issue:9727",94        "issue:6363|issue:7330",95        "issue:2780|issue:6363",96        "issue:4104|issue:6815",97        "issue:11169|issue:6517",98        "issue:11417|issue:2044"99      ],100      "prepared_review_unit_hash": "0133da4827865cc18ff504a8d40a95b1bf43e4a1d70288f3555e8f0c674ee8d9",101      "component_budget": {102        "node_count": 3014,103        "item_count": 3014,104        "soft_pair_count": 8931,105        "serialized_chars": 4765605,106        "estimated_input_tokens": 1191402,107        "estimated_eval_tokens": 2383060108      },109      "budget": {110        "node_count": 48,111        "item_count": 48,112        "soft_pair_count": 28,113        "serialized_chars": 36481,114        "estimated_input_tokens": 9121,115        "estimated_eval_tokens": 18498116      },117      "overflow_policy": "truncate_then_skip",118      "trimmed": false,119      "aggressively_trimmed": false,120      "split": true,121      "analyst_result": {122        "summary": "The strongest duplicate clusters are the exact-title or near-exact error repeats, especially the CLIPTextModel/offload_state_dict reports. I kept broader symptom matches conservative when the titles point to different model or pipeline contexts and the packet lacks body text or traces to prove a shared root cause.",123        "confidence": 0.76,124        "canonical_issue_reason": "issue:12480 is the cleanest anchor for the offload_state_dict/CLIPTextModel cluster: it has the most complete wording, and 12481 plus 12436 are very close variants of the same report.",125        "canonical_pr_reason": null,126        "best_issue_reason": "issue:12480 is the best overall issue anchor because it is the clearest, most detailed representative of the strongest duplicate cluster.",127        "best_pr_reason": null,128        "soft_edge_verdicts": [129          {130            "left": "issue:12480",131            "right": "issue:12481",132            "accept": true,133            "reason": "Near-identical titles and the same CLIPTextModel/offload_state_dict TypeError; same bug."134          },135          {136            "left": "issue:5295",137            "right": "issue:5475",138            "accept": true,139            "reason": "Same LaVie/LaVIE project title with only capitalization variation; duplicate report."140          },141          {142            "left": "issue:10035",143            "right": "issue:10091",144            "accept": true,145            "reason": "Exact same CogvideoX VAE decoder memory issue title; duplicate."146          },147          {148            "left": "issue:5510",149            "right": "issue:7854",150            "accept": false,151            "reason": "Both mention dtype mismatches, but the pipeline/context differs and the packet does not show one concrete shared code path."152          },153          {154            "left": "issue:3954",155            "right": "issue:8236",156            "accept": false,157            "reason": "Same broad scheduler step-count symptom, but not enough shared evidence here to confirm the same underlying bug."158          },159          {160            "left": "issue:12436",161            "right": "issue:12480",162            "accept": true,163            "reason": "Same offload_state_dict / CLIPTextModel TypeError family; clearly overlapping reports."164          },165          {166            "left": "issue:12436",167            "right": "issue:12481",168            "accept": true,169            "reason": "Same offload_state_dict / CLIPTextModel TypeError family; duplicate variant."170          },171          {172            "left": "issue:6815",173            "right": "issue:8759",174            "accept": false,175            "reason": "Both are black-image / invalid-cast symptom reports, but the model context differs and the packet does not prove the same root cause."176          },177          {178            "left": "issue:5805",179            "right": "issue:7854",180            "accept": false,181            "reason": "Related dtype mismatch wording, but the specific failing path and context differ too much to merge them confidently."182          },183          {184            "left": "issue:10804",185            "right": "issue:10954",186            "accept": true,187            "reason": "Same FLUX1.dev LoRA load-failure report, just reworded."188          },189          {190            "left": "issue:4325",191            "right": "issue:8759",192            "accept": false,193            "reason": "The error text is similar, but this is still only a broad black-image symptom match across different model contexts."194          },195          {196            "left": "issue:4754",197            "right": "issue:5259",198            "accept": true,199            "reason": "Same missing model_index.json / textual inversion loading error; duplicate."200          },201          {202            "left": "issue:6727",203            "right": "issue:6812",204            "accept": true,205            "reason": "Same Query/Key/Value dtype mismatch during LoRA training; same bug family."206          },207          {208            "left": "issue:4649",209            "right": "issue:6996",210            "accept": true,211            "reason": "Same NoneType / added_cond_kwargs text_embeds error; duplicate."212          },213          {214            "left": "issue:11083",215            "right": "issue:3809",216            "accept": true,217            "reason": "Same torch-DDP stride mismatch warning; near-exact duplicate wording."218          },219          {220            "left": "issue:12044",221            "right": "issue:6324",222            "accept": true,223            "reason": "Exact same AttributeError about bool.__module__; duplicate."224          },225          {226            "left": "issue:10343",227            "right": "issue:6815",228            "accept": false,229            "reason": "Both mention black output / invalid-cast symptoms, but they are model-specific reports and not clearly the same bug."230          },231          {232            "left": "issue:2995",233            "right": "issue:5082",234            "accept": true,235            "reason": "Same mat1/mat2 shape-mismatch error; duplicate."236          },237          {238            "left": "issue:2819",239            "right": "issue:6594",240            "accept": false,241            "reason": "Both are raw.githubusercontent.com download failures, but the target files and underlying cause are not shown to be the same."242          },243          {244            "left": "issue:9293",245            "right": "issue:9298",246            "accept": true,247            "reason": "Exact same Float8_e4m3fnStorage lookup failure; duplicate."248          },249          {250            "left": "issue:4557",251            "right": "issue:4868",252            "accept": true,253            "reason": "Same super object __getattr__ AttributeError; duplicate."254          },255          {256            "left": "issue:11225",257            "right": "issue:9546",258            "accept": false,259            "reason": "Both are validation-step OOM reports, but they are model-family-specific and only symptom-level similar here."260          },261          {262            "left": "issue:9726",263            "right": "issue:9727",264            "accept": false,265            "reason": "Related FLUX.1-dev multi-GPU reports, but one is a training problem and the other a save problem; not the same bug."266          },267          {268            "left": "issue:6363",269            "right": "issue:7330",270            "accept": true,271            "reason": "Same Attempting to unscale FP16 gradients error in the LoRA training script family; duplicate."272          },273          {274            "left": "issue:2780",275            "right": "issue:6363",276            "accept": true,277            "reason": "Same FP16 unscale failure in text-to-image LoRA training; duplicate."278          },279          {280            "left": "issue:4104",281            "right": "issue:6815",282            "accept": false,283            "reason": "Same broad black-image/invalid-cast symptom, but not enough evidence of one shared root cause across the different reports."284          },285          {286            "left": "issue:11169",287            "right": "issue:6517",288            "accept": false,289            "reason": "Both are import failures, but the reported errors differ and do not indicate the same bug."290          },291          {292            "left": "issue:11417",293            "right": "issue:2044",294            "accept": true,295            "reason": "Exact same DistributedDataParallel has no dtype AttributeError; duplicate."296          }297        ]298      },299      "evaluator_result": {300        "accept": false,301        "feedback": "The canonical choice is reasonably grounded, and several exact-title or near-exact duplicate pairs are fine. However, the soft-edge section is too assertive in a few places given the packet only provides titles/body sizes, not body text or code traces. In particular, pairs like issue:11225 vs issue:9546, issue:6815 vs issue:8759, issue:10343 vs issue:6815, and issue:3954 vs issue:8236 read like symptom-level similarity rather than clearly the same bug. For this task, those should stay conservative unless the packet shows stronger shared evidence. The summary should also avoid implying direct proof of sameness for broad symptom matches when the model/pipeline context differs."302      },303      "evaluator_used": true,304      "retried": true,305      "accepted_nontrivial_soft_edge": true,306      "error_kind": null,307      "error_message": null308    },309    {310      "label": "issue",311      "component_index": 1,312      "component_count": 102,313      "review_unit_index": 2,314      "review_unit_count": 364,315      "status": "reviewed",316      "reason": null,317      "source": "cache",318      "cache_hit": true,319      "model": "gpt-5.4-mini?service_tier=flex",320      "cluster_id": "cluster-50-3014",321      "nodes": [322        "issue:10520",323        "issue:10575",324        "issue:10650",325        "issue:10673",326        "issue:10958",327        "issue:11413",328        "issue:11436",329        "issue:12009",330        "issue:12104",331        "issue:12559",332        "issue:1433",333        "issue:1875",334        "issue:2500",335        "issue:2780",336        "issue:3489",337        "issue:5110",338        "issue:5510",339        "issue:5607",340        "issue:5749",341        "issue:5805",342        "issue:5897",343        "issue:6086",344        "issue:6363",345        "issue:6368",346        "issue:6445",347        "issue:6517",348        "issue:6570",349        "issue:6821",350        "issue:6832",351        "issue:6879",352        "issue:7261",353        "issue:7286",354        "issue:7330",355        "issue:736",356        "issue:7681",357        "issue:7708",358        "issue:7854",359        "issue:7891",360        "issue:7935",361        "issue:7978",362        "issue:8465",363        "issue:8581",364        "issue:8833",365        "issue:8863",366        "issue:9362",367        "issue:9448",368        "issue:9476"369      ],370      "soft_pairs": [371        "issue:6517|issue:7681",372        "issue:5749|issue:9362",373        "issue:12559|issue:7978",374        "issue:5607|issue:6570",375        "issue:7935|issue:8833",376        "issue:12104|issue:9362",377        "issue:1875|issue:5510",378        "issue:10650|issue:10673",379        "issue:7854|issue:9476",380        "issue:11413|issue:6517",381        "issue:7708|issue:7935",382        "issue:5110|issue:736",383        "issue:5897|issue:7891",384        "issue:6086|issue:6363",385        "issue:7261|issue:7286",386        "issue:1875|issue:7854",387        "issue:11436|issue:6879",388        "issue:10958|issue:12009",389        "issue:10958|issue:2500",390        "issue:8863|issue:9448",391        "issue:1875|issue:5805",392        "issue:6445|issue:6821",393        "issue:2780|issue:7330",394        "issue:6832|issue:8863",395        "issue:7935|issue:8465",396        "issue:10520|issue:7891",397        "issue:6368|issue:8581",398        "issue:10575|issue:5749",399        "issue:1433|issue:3489"400      ],401      "prepared_review_unit_hash": "85ad060b999b6dfe4d5e443055781096e05483f557cb1d6f1af4713b8d076769",402      "component_budget": {403        "node_count": 3014,404        "item_count": 3014,405        "soft_pair_count": 8931,406        "serialized_chars": 4765605,407        "estimated_input_tokens": 1191402,408        "estimated_eval_tokens": 2383060409      },410      "budget": {411        "node_count": 47,412        "item_count": 47,413        "soft_pair_count": 29,414        "serialized_chars": 36851,415        "estimated_input_tokens": 9213,416        "estimated_eval_tokens": 18682417      },418      "overflow_policy": "truncate_then_skip",419      "trimmed": false,420      "aggressively_trimmed": false,421      "split": true,422      "analyst_result": {423        "summary": "The cluster still breaks into a few well-supported duplicate subgroups, especially the LoRA FP16 unscale reports and a handful of exact-message matches. However, several tuple.shape links are too weakly grounded to treat as confirmed duplicates, so I\u2019d keep the triage conservative and not collapse the broader cluster too aggressively.",424        "confidence": 0.69,425        "canonical_issue_reason": "Issue 6086 is the strongest canonical anchor for the accepted duplicate subgroup: it has the richest discussion, a clear repro, and matches the same FP16-gradient unscale failure seen in 6363 and 7330.",426        "canonical_pr_reason": null,427        "best_issue_reason": "Issue 6086 is the best overall issue to anchor the confirmed subgroup because it is detailed, well-scoped, and explicitly captures the shared LoRA training failure mode.",428        "best_pr_reason": null,429        "soft_edge_verdicts": [430          {431            "left": "issue:6517",432            "right": "issue:7681",433            "accept": false,434            "reason": "Both are import-time failures, but they involve different modules and different underlying causes."435          },436          {437            "left": "issue:5749",438            "right": "issue:9362",439            "accept": false,440            "reason": "Both are IndexErrors, but the reported sizes differ and there is no clear shared failing code path."441          },442          {443            "left": "issue:12559",444            "right": "issue:7978",445            "accept": true,446            "reason": "Same concrete torchvision::nms missing-operator failure; the later report is just a more specific instance of the same bug."447          },448          {449            "left": "issue:5607",450            "right": "issue:6570",451            "accept": true,452            "reason": "Both report the same tensor-size mismatch in the StableDiffusionXLInstructPix2Pix pipeline family."453          },454          {455            "left": "issue:7935",456            "right": "issue:8833",457            "accept": false,458            "reason": "Both mention missing config.json, but they point to different models and loader contexts."459          },460          {461            "left": "issue:12104",462            "right": "issue:9362",463            "accept": false,464            "reason": "Same broad IndexError wording, but the tensor sizes and failure context are clearly different."465          },466          {467            "left": "issue:1875",468            "right": "issue:5510",469            "accept": false,470            "reason": "Both are dtype mismatch errors, but the pipeline context and concrete failure mode are not the same."471          },472          {473            "left": "issue:10650",474            "right": "issue:10673",475            "accept": false,476            "reason": "The shape text matches, but the pipelines/models differ enough that this is not clearly the same underlying bug."477          },478          {479            "left": "issue:7854",480            "right": "issue:9476",481            "accept": false,482            "reason": "Both are dtype mismatch reports, but they arise in different training/validation contexts and are too generic to merge."483          },484          {485            "left": "issue:11413",486            "right": "issue:6517",487            "accept": false,488            "reason": "Both are import failures, but the traceback targets different pipelines and does not indicate a shared root cause."489          },490          {491            "left": "issue:7708",492            "right": "issue:7935",493            "accept": false,494            "reason": "Both involve missing config.json, but the model/path context differs and the evidence is too weak for duplication."495          },496          {497            "left": "issue:5110",498            "right": "issue:736",499            "accept": true,500            "reason": "Same HFValidationError about invalid repo IDs; the second report is the same validation bug in another form."501          },502          {503            "left": "issue:5897",504            "right": "issue:7891",505            "accept": false,506            "reason": "The device-mismatch wording is similar, but the reports are too context-poor to prove the same concrete bug."507          },508          {509            "left": "issue:6086",510            "right": "issue:6363",511            "accept": true,512            "reason": "Same exact FP16 unscale failure in the LoRA text-to-image training script family."513          },514          {515            "left": "issue:7261",516            "right": "issue:7286",517            "accept": true,518            "reason": "Both are unsuccessful cross-attention weight loading reports for Custom Diffusion, with the second adding a pipeline variant."519          },520          {521            "left": "issue:1875",522            "right": "issue:7854",523            "accept": false,524            "reason": "Both are half/float mismatch reports, but the surrounding pipelines differ and the bug is not concretely the same."525          },526          {527            "left": "issue:11436",528            "right": "issue:6879",529            "accept": false,530            "reason": "Both are mat1/mat2 dtype mismatch errors, but one is half vs bfloat16 and the other is half vs float."531          },532          {533            "left": "issue:10958",534            "right": "issue:12009",535            "accept": false,536            "reason": "Both mention meta-tensor loading problems, but the model format and loader path differ enough to keep them separate."537          },538          {539            "left": "issue:10958",540            "right": "issue:2500",541            "accept": false,542            "reason": "Same meta-tensor symptom, but one is a specific pipeline loader report and the other is test failures; too broad to merge."543          },544          {545            "left": "issue:8863",546            "right": "issue:9448",547            "accept": false,548            "reason": "The tuple.shape symptom is similar, but the evidence is only a generic exception string plus a broad pipeline family, so this is not confirmed duplicate-level evidence."549          },550          {551            "left": "issue:1875",552            "right": "issue:5805",553            "accept": false,554            "reason": "Both are dtype-related, but the titles do not show the same concrete failure path."555          },556          {557            "left": "issue:6445",558            "right": "issue:6821",559            "accept": false,560            "reason": "Both are caption_column validation issues, but the accepted values and dataset-schema contexts differ."561          },562          {563            "left": "issue:2780",564            "right": "issue:7330",565            "accept": true,566            "reason": "Same exact Attempting to unscale FP16 gradients failure in the LoRA text-to-image training script family."567          },568          {569            "left": "issue:6832",570            "right": "issue:8863",571            "accept": false,572            "reason": "This is only supported by a generic tuple.shape exception and broad pipeline-family similarity, which is not enough to confirm a duplicate."573          },574          {575            "left": "issue:7935",576            "right": "issue:8465",577            "accept": false,578            "reason": "Both mention missing config.json, but they involve different pipelines and model-loading paths."579          },580          {581            "left": "issue:10520",582            "right": "issue:7891",583            "accept": false,584            "reason": "Same device-mismatch wording, but the reports are too vague to prove the same code path."585          },586          {587            "left": "issue:6368",588            "right": "issue:8581",589            "accept": false,590            "reason": "Both are unsupported-lora-layer errors, but they reference different layers and are not the same concrete bug."591          },592          {593            "left": "issue:10575",594            "right": "issue:5749",595            "accept": false,596            "reason": "Both are index-out-of-bounds errors, but the dimensions and sizes differ and there is no clear shared failure path."597          },598          {599            "left": "issue:1433",600            "right": "issue:3489",601            "accept": true,602            "reason": "Same OnnxRuntimeModel has no attribute config error; the second is a pipeline-specific duplicate of the first."603          }604        ]605      },606      "evaluator_result": {607        "accept": false,608        "feedback": "The summary is mostly grounded, but a few soft-edge accepts are too strong for the evidence in the packet. In particular, the tuple.shape links (issue:6832 \u2194 issue:8863 and issue:8863 \u2194 issue:9448) are only supported by a generic exception string plus a broad pipeline family, so treating them as confirmed duplicates overstates the packet. The LoRA FP16-unscale subgroup and the exact-message pairs (e.g. HFValidationError, torchvision::nms, OnnxRuntimeModel.config) are better supported. Because the output accepts some weakly grounded merges, I\u2019d reject for precision."609      },610      "evaluator_used": true,611      "retried": true,612      "accepted_nontrivial_soft_edge": true,613      "error_kind": null,614      "error_message": null615    },616    {617      "label": "issue",618      "component_index": 1,619      "component_count": 102,620      "review_unit_index": 3,621      "review_unit_count": 364,622      "status": "reviewed",623      "reason": null,624      "source": "cache",625      "cache_hit": true,626      "model": "gpt-5.4-mini?service_tier=flex",627      "cluster_id": "cluster-50-3014",628      "nodes": [629        "issue:10520",630        "issue:10722",631        "issue:10896",632        "issue:10897",633        "issue:10952",634        "issue:11436",635        "issue:12436",636        "issue:13261",637        "issue:13311",638        "issue:3357",639        "issue:3651",640        "issue:4410",641        "issue:4752",642        "issue:4932",643        "issue:4933",644        "issue:5022",645        "issue:5025",646        "issue:5399",647        "issue:5510",648        "issue:5603",649        "issue:5661",650        "issue:6124",651        "issue:6278",652        "issue:6363",653        "issue:6482",654        "issue:7003",655        "issue:7330",656        "issue:7708",657        "issue:7978",658        "issue:8518",659        "issue:859",660        "issue:8833",661        "issue:9063",662        "issue:9109",663        "issue:9291",664        "issue:9292",665        "issue:9443",666        "issue:9448",667        "issue:9602",668        "issue:9628",669        "issue:9726",670        "issue:9727",671        "issue:9790",672        "issue:9799",673        "issue:9895",674        "issue:9901",675        "issue:9973"676      ],677      "soft_pairs": [678        "issue:10896|issue:10897",679        "issue:9063|issue:9443",680        "issue:4932|issue:4933",681        "issue:3651|issue:6278",682        "issue:9291|issue:9292",683        "issue:5022|issue:5661",684        "issue:10722|issue:9799",685        "issue:9727|issue:9790",686        "issue:6363|issue:6482",687        "issue:10952|issue:9602",688        "issue:9726|issue:9790",689        "issue:3357|issue:8518",690        "issue:7330|issue:9628",691        "issue:6124|issue:7330",692        "issue:9628|issue:9973",693        "issue:9448|issue:9901",694        "issue:7708|issue:8833",695        "issue:10520|issue:9895",696        "issue:13261|issue:13311",697        "issue:11436|issue:5399",698        "issue:7978|issue:9109",699        "issue:5025|issue:5603",700        "issue:12436|issue:859",701        "issue:4752|issue:5510",702        "issue:4410|issue:7003"703      ],704      "prepared_review_unit_hash": "607f7cefb8a8b1a8a7a41bf1917a6d6b32b91415c039c9b64fcb91cb7e765f87",705      "component_budget": {706        "node_count": 3014,707        "item_count": 3014,708        "soft_pair_count": 8931,709        "serialized_chars": 4765605,710        "estimated_input_tokens": 1191402,711        "estimated_eval_tokens": 2383060712      },713      "budget": {714        "node_count": 47,715        "item_count": 47,716        "soft_pair_count": 25,717        "serialized_chars": 34728,718        "estimated_input_tokens": 8682,719        "estimated_eval_tokens": 17620720      },721      "overflow_policy": "truncate_then_skip",722      "trimmed": false,723      "aggressively_trimmed": false,724      "split": true,725      "analyst_result": {726        "summary": "The set is heterogeneous overall, but many pairs are clear duplicate issues around the same error or request. I accepted only pairs that appear to describe the same underlying bug/feature and rejected pairs that merely share a subsystem or broad error class.",727        "confidence": 0.89,728        "canonical_issue_reason": "No single canonical issue covers the full set because the items split into many unrelated subclusters. The cleanest duplicate anchor is issue 10897, which is an exact same-title match to 10896 and remains open.",729        "canonical_pr_reason": null,730        "best_issue_reason": "Issue 10897 is the best issue to keep as the canonical representative for its subcluster: it is an exact duplicate of 10896, has the same request text, and is open.",731        "best_pr_reason": null,732        "soft_edge_verdicts": [733          {734            "left": "issue:10896",735            "right": "issue:10897",736            "accept": true,737            "reason": "Exact same title and same feature request; these are duplicates."738          },739          {740            "left": "issue:9063",741            "right": "issue:9443",742            "accept": false,743            "reason": "Same error string, but different pipelines/models; likely separate tokenizer-loading cases."744          },745          {746            "left": "issue:4932",747            "right": "issue:4933",748            "accept": true,749            "reason": "Same DreamBooth output complaint with near-identical wording; duplicate reports."750          },751          {752            "left": "issue:3651",753            "right": "issue:6278",754            "accept": false,755            "reason": "Both mention multi-GPU, but they affect different training scripts and likely different failure points."756          },757          {758            "left": "issue:9291",759            "right": "issue:9292",760            "accept": true,761            "reason": "Both report FLUX LoRA key incompatibility with diffusers loaders; same underlying compatibility bug."762          },763          {764            "left": "issue:5022",765            "right": "issue:5661",766            "accept": true,767            "reason": "Both are the same missing PreTrainedTokenizer import/name error in diffusers."768          },769          {770            "left": "issue:10722",771            "right": "issue:9799",772            "accept": false,773            "reason": "Both are tensor-size mismatches, but the actual dimensions and training contexts differ."774          },775          {776            "left": "issue:9727",777            "right": "issue:9790",778            "accept": false,779            "reason": "One is a save problem and the other is a training problem; not clearly the same bug."780          },781          {782            "left": "issue:6363",783            "right": "issue:6482",784            "accept": true,785            "reason": "Same FP16 unscale error in closely related DreamBooth/LoRA training paths."786          },787          {788            "left": "issue:10952",789            "right": "issue:9602",790            "accept": true,791            "reason": "Both are adapter-name lookup failures during LoRA loading; same adapter handling issue."792          },793          {794            "left": "issue:9726",795            "right": "issue:9790",796            "accept": true,797            "reason": "Same FLUX.1-dev DreamBooth multigpu training failure theme and likely same code path."798          },799          {800            "left": "issue:3357",801            "right": "issue:8518",802            "accept": false,803            "reason": "Both are torch.compile errors, but they are different unsupported operations and different stack paths."804          },805          {806            "left": "issue:7330",807            "right": "issue:9628",808            "accept": true,809            "reason": "Same \"Attempting to unscale FP16 gradients\" training failure across related scripts."810          },811          {812            "left": "issue:6124",813            "right": "issue:7330",814            "accept": true,815            "reason": "Same FP16 gradient unscale error in DreamBooth/LoRA training flows."816          },817          {818            "left": "issue:9628",819            "right": "issue:9973",820            "accept": true,821            "reason": "Both report the same FP16 unscale failure in DreamBooth LoRA SD3 training."822          },823          {824            "left": "issue:9448",825            "right": "issue:9901",826            "accept": true,827            "reason": "Both describe the same tuple-vs-shape attention/processor bug path."828          },829          {830            "left": "issue:7708",831            "right": "issue:8833",832            "accept": false,833            "reason": "Both are missing-file model loading errors, but they concern different repos and different missing files."834          },835          {836            "left": "issue:10520",837            "right": "issue:9895",838            "accept": false,839            "reason": "Same device-mismatch symptom, but different models and no evidence they share the same fix."840          },841          {842            "left": "issue:13261",843            "right": "issue:13311",844            "accept": true,845            "reason": "Both are FLUX.2-klein-9B LoRA compatibility failures in the same transformer/LoRA loading path."846          },847          {848            "left": "issue:11436",849            "right": "issue:5399",850            "accept": false,851            "reason": "Both are dtype mismatches, but they involve different pipelines and are too generic to merge confidently."852          },853          {854            "left": "issue:7978",855            "right": "issue:9109",856            "accept": true,857            "reason": "Exact same torchvision::nms missing-operator error."858          },859          {860            "left": "issue:5025",861            "right": "issue:5603",862            "accept": false,863            "reason": "Both are import errors from diffusers.utils, but they are different missing symbols and likely separate breakages."864          },865          {866            "left": "issue:12436",867            "right": "issue:859",868            "accept": false,869            "reason": "Both are unexpected-argument constructor errors, but for different classes and different arguments."870          },871          {872            "left": "issue:4752",873            "right": "issue:5510",874            "accept": false,875            "reason": "Both are dtype mismatches in SDXL-family pipelines, but the concrete pipelines and fixes are different."876          },877          {878            "left": "issue:4410",879            "right": "issue:7003",880            "accept": false,881            "reason": "Both involve weight-shape mismatches, but they are different in/out channel expectations and different loading scenarios."882          }883        ]884      },885      "evaluator_result": {886        "accept": true,887        "feedback": "The summary is grounded in the packet and stays conservative overall. The canonical anchor choice is supported by the identical SkyReels-A1 titles and open state on 10897, and the accepted soft-edge pairs all have reasonably close title/body alignment rather than just broad subsystem similarity. The rejected pairs are also appropriately cautious. No clear overstatement of evidence stood out."888      },889      "evaluator_used": true,890      "retried": false,891      "accepted_nontrivial_soft_edge": true,892      "error_kind": null,893      "error_message": null894    },895    {896      "label": "issue",897      "component_index": 1,898      "component_count": 102,899      "review_unit_index": 4,900      "review_unit_count": 364,901      "status": "reviewed",902      "reason": null,903      "source": "cache",904      "cache_hit": true,905      "model": "gpt-5.4-mini?service_tier=flex",906      "cluster_id": "cluster-50-3014",907      "nodes": [908        "issue:10154",909        "issue:10343",910        "issue:10467",911        "issue:10928",912        "issue:11218",913        "issue:11410",914        "issue:11417",915        "issue:11702",916        "issue:11709",917        "issue:11849",918        "issue:11865",919        "issue:11901",920        "issue:12104",921        "issue:12174",922        "issue:12778",923        "issue:13261",924        "issue:13350",925        "issue:1925",926        "issue:1931",927        "issue:1946",928        "issue:1976",929        "issue:2100",930        "issue:2947",931        "issue:3325",932        "issue:3481",933        "issue:3829",934        "issue:4265",935        "issue:4752",936        "issue:4868",937        "issue:5259",938        "issue:5510",939        "issue:5805",940        "issue:6139",941        "issue:6363",942        "issue:6821",943        "issue:6914",944        "issue:7049",945        "issue:720",946        "issue:7210",947        "issue:7708",948        "issue:7851",949        "issue:7854",950        "issue:8759",951        "issue:8975",952        "issue:9210",953        "issue:9628",954        "issue:9690",955        "issue:9901"956      ],957      "soft_pairs": [958        "issue:10154|issue:7210",959        "issue:1931|issue:720",960        "issue:4752|issue:7854",961        "issue:11417|issue:12778",962        "issue:1946|issue:3325",963        "issue:6363|issue:9628",964        "issue:6914|issue:9901",965        "issue:13261|issue:13350",966        "issue:1925|issue:1931",967        "issue:2947|issue:6139",968        "issue:10928|issue:5259",969        "issue:10343|issue:8759",970        "issue:5510|issue:5805",971        "issue:4752|issue:5805",972        "issue:12104|issue:7851",973        "issue:10467|issue:9690",974        "issue:6821|issue:9210",975        "issue:12174|issue:4868",976        "issue:3481|issue:3829",977        "issue:11417|issue:11865",978        "issue:11218|issue:8975",979        "issue:11702|issue:11709",980        "issue:1976|issue:7708",981        "issue:2100|issue:4265",982        "issue:11410|issue:7049",983        "issue:11849|issue:11901"984      ],985      "prepared_review_unit_hash": "73578b402a4ad4b2104ce6c628216c4a575d394ab066dcd47fccc3a7c973f6e2",986      "component_budget": {987        "node_count": 3014,988        "item_count": 3014,989        "soft_pair_count": 8931,990        "serialized_chars": 4765605,991        "estimated_input_tokens": 1191402,992        "estimated_eval_tokens": 2383060993      },994      "budget": {995        "node_count": 48,996        "item_count": 48,997        "soft_pair_count": 26,998        "serialized_chars": 35606,999        "estimated_input_tokens": 8902,1000        "estimated_eval_tokens": 180601001      },1002      "overflow_policy": "truncate_then_skip",1003      "trimmed": false,1004      "aggressively_trimmed": false,1005      "split": true,1006      "analyst_result": {1007        "summary": "This set is not one tight duplicate cluster; it contains several separate bug families. Clear duplicate-like pairs mostly involve repeated dtype-mismatch errors, repeated missing-file/import lookups, repeated attribute errors, and repeated LoRA key-loading failures. Several other soft pairs are only broadly similar and should stay separate.",1008        "confidence": 0.84,1009        "canonical_issue_reason": "No single canonical issue covers the whole set; the items split into multiple unrelated bug families. If one representative must be chosen, issue 1931 best captures the recurring half-precision dtype-mismatch subgroup.",1010        "canonical_pr_reason": null,1011        "best_issue_reason": "Issue 1931 is the strongest overall representative because it is a clear, recurring half-precision/float dtype-mismatch report and aligns with other near-duplicate reports in that subgroup.",1012        "best_pr_reason": null,1013        "soft_edge_verdicts": [1014          {1015            "left": "issue:10154",1016            "right": "issue:7210",1017            "accept": false,1018            "reason": "Both are AttributeError reports, but they hit different APIs and code paths: load_lora_weights vs accelerate.load_checkpoint_and_dispatch. Not the same bug."1019          },1020          {1021            "left": "issue:1931",1022            "right": "issue:720",1023            "accept": true,1024            "reason": "Same underlying float16/Half vs Float mismatch during Stable Diffusion inference; the symptom and fix path are effectively the same."1025          },1026          {1027            "left": "issue:4752",1028            "right": "issue:7854",1029            "accept": true,1030            "reason": "These are the same input/bias dtype mismatch error in SDXL code, with nearly identical wording and likely the same fix."1031          },1032          {1033            "left": "issue:11417",1034            "right": "issue:12778",1035            "accept": false,1036            "reason": "Both are missing-attribute errors, but on different objects and different failures. Too little evidence of one concrete shared bug."1037          },1038          {1039            "left": "issue:1946",1040            "right": "issue:3325",1041            "accept": false,1042            "reason": "Both mention CUDA invalid argument, but one is xformers-related and the other is train_dreambooth.py. Same broad symptom, not the same bug."1043          },1044          {1045            "left": "issue:6363",1046            "right": "issue:9628",1047            "accept": false,1048            "reason": "Same error text, but different training scripts and model setups. Insufficient evidence that one code-path fix would address both."1049          },1050          {1051            "left": "issue:6914",1052            "right": "issue:9901",1053            "accept": false,1054            "reason": "Both are tuple/shape-related AttributeErrors, but they occur in different contexts (IP-Adapter/cpu offload vs AttnProcessor) and do not clearly share one root cause."1055          },1056          {1057            "left": "issue:13261",1058            "right": "issue:13350",1059            "accept": false,1060            "reason": "One is a LoRA/lokr loading failure; the other is a batch inference usage question. Not the same underlying bug."1061          },1062          {1063            "left": "issue:1925",1064            "right": "issue:1931",1065            "accept": true,1066            "reason": "Both describe the same half-precision/float mismatch class of bug in Stable Diffusion inference."1067          },1068          {1069            "left": "issue:2947",1070            "right": "issue:6139",1071            "accept": false,1072            "reason": "These are different missing attributes on UNet2DConditionModel, so they are not the same concrete defect."1073          },1074          {1075            "left": "issue:10928",1076            "right": "issue:5259",1077            "accept": false,1078            "reason": "Different missing-file errors: fp16 weight file vs model_index.json. Separate problems."1079          },1080          {1081            "left": "issue:10343",1082            "right": "issue:8759",1083            "accept": true,1084            "reason": "Same invalid-cast warning in image postprocessing leading to black images; the symptom and code area match closely."1085          },1086          {1087            "left": "issue:5510",1088            "right": "issue:5805",1089            "accept": false,1090            "reason": "Both are dtype mismatch errors, but they occur in different pipelines/platforms and the exact failure context differs too much to merge confidently."1091          },1092          {1093            "left": "issue:4752",1094            "right": "issue:5805",1095            "accept": false,1096            "reason": "Same broad dtype mismatch theme, but one is an SDXL pipeline bug and the other is an XLA-specific issue with opposite type ordering. Not a clean duplicate."1097          },1098          {1099            "left": "issue:12104",1100            "right": "issue:7851",1101            "accept": false,1102            "reason": "Both are index-out-of-bounds errors, but one is size 0 and the other is a concurrency-related size 11 case. Different bugs."1103          },1104          {1105            "left": "issue:10467",1106            "right": "issue:9690",1107            "accept": false,1108            "reason": "One is about a generated temp filename in FP8 example output; the other is a cleanup request. Not the same issue."1109          },1110          {1111            "left": "issue:6821",1112            "right": "issue:9210",1113            "accept": false,1114            "reason": "These are opposite dataset-column validation errors (caption_column vs image_column) and do not indicate the same defect."1115          },1116          {1117            "left": "issue:12174",1118            "right": "issue:4868",1119            "accept": true,1120            "reason": "Same AttributeError on __getattr__ with the same underlying failure signature."1121          },1122          {1123            "left": "issue:3481",1124            "right": "issue:3829",1125            "accept": false,1126            "reason": "Both discuss image differences across inference settings, but they involve different schedulers/pipelines and are too broad to treat as one bug."1127          },1128          {1129            "left": "issue:11417",1130            "right": "issue:11865",1131            "accept": false,1132            "reason": "Completely different attribute errors on different objects; no shared code-path is apparent."1133          },1134          {1135            "left": "issue:11218",1136            "right": "issue:8975",1137            "accept": false,1138            "reason": "Both are import errors from diffusers.loaders, but they concern different symbols and likely different missing exports."1139          },1140          {1141            "left": "issue:11702",1142            "right": "issue:11709",1143            "accept": true,1144            "reason": "Both are LoRA key-loading failures in diffusers for closely related video-model use cases; a single loader fix could plausibly address both."1145          },1146          {1147            "left": "issue:1976",1148            "right": "issue:7708",1149            "accept": true,1150            "reason": "Same incorrect tokenizer/config.json file lookup against a model repo that does not provide it."1151          },1152          {1153            "left": "issue:2100",1154            "right": "issue:4265",1155            "accept": false,1156            "reason": "Different missing imports for different pipelines/classes; same category, but not the same concrete bug."1157          },1158          {1159            "left": "issue:11410",1160            "right": "issue:7049",1161            "accept": true,1162            "reason": "Exact same too-many-values-to-unpack error in the attention/IP-adapter code path."1163          }1164        ]1165      },1166      "evaluator_result": {1167        "accept": true,1168        "feedback": "The summary is grounded in the packet and stays conservative overall. The accepted soft-edge pairs are mostly supported by near-identical error signatures/titles, and the rejected pairs avoid overclaiming when only broad symptom overlap is present. No clear overstatement stands out."1169      },1170      "evaluator_used": true,1171      "retried": false,1172      "accepted_nontrivial_soft_edge": true,1173      "error_kind": null,1174      "error_message": null1175    },1176    {1177      "label": "issue",1178      "component_index": 1,1179      "component_count": 102,1180      "review_unit_index": 5,1181      "review_unit_count": 364,1182      "status": "reviewed",1183      "reason": null,1184      "source": "cache",1185      "cache_hit": true,1186      "model": "gpt-5.4-mini?service_tier=flex",1187      "cluster_id": "cluster-50-3014",1188      "nodes": [1189        "issue:10154",1190        "issue:10472",1191        "issue:11218",1192        "issue:12174",1193        "issue:12436",1194        "issue:12480",1195        "issue:12481",1196        "issue:12482",1197        "issue:12719",1198        "issue:12722",1199        "issue:12895",1200        "issue:1709",

Showing the first 1,200 of 209387 lines. Download the file for the rest.