A Pocket-Sized MLLM for Ultra-Efficient Image and Video Understanding on Your Phone
This report summarizes scanner evidence for OpenBMB/MiniCPM-V, a Python project with 26,530 GitHub stars. SynthScan examined 58,262 lines of code across 225 source files, recording 641 pattern matches distributed across 16 syntactic categories. The legacy adjusted pattern signal is 13.1 (Moderate pattern signal); it is not an authorship estimate.
Use the findings to ask whether maintainers can explain intent, verify dependencies, simplify unnecessary scaffolding, and focus review effort. Risk Payload v1 adds ownership, supply-chain, over-engineering, and review-effort vectors when available. These are prompts for human verification, not probabilities, security findings, or measured maintenance costs.
Use each vector to guide manual verification of the underlying findings.
Can maintainers explain intent, assumptions, and trade-offs? 264 finding(s).
Are imports declared and package/API names verified? 258 finding(s).
Are wrappers, guards, and repeated structures necessary? 208 finding(s).
Which changes need focused human review? 641 finding(s).
This chart tracks the legacy adjusted pattern score across scan runs. Changes can reflect source changes, scan coverage, or detector/scoring versions; they do not measure AI adoption or prove that a team refactored code. Compare runs only when their scoring versions are compatible. The dashed line tracks the raw pattern-hit count.
Groups findings by configured severity weight. Severity is not a confidence estimate, security rating, or proof that a package is missing. Verify imports against the project's manifests and inspect each finding in its local design context.
This horizontal bar chart shows where configured pattern scores accumulate across directories. Treat the bars as locations to inspect, not as evidence of authorship or code quality. Use the file-level findings to decide whether a directory needs a focused ownership, dependency, complexity, or review check.
The scanner identified 641 distinct pattern matches across 16 syntactic categories. Each entry below represents a discrete location in the source code where the engine recorded a detector match to inspect in context. Expand any category row to inspect the individual file paths, line numbers, code snippets, and the lexical context (CODE, COMMENT, or STRING) in which each match was detected.
Reading the findings table: The Severity column indicates the configured finding weight (CRITICAL / HIGH / MEDIUM / LOW). The Context column identifies whether the match occurred inside executable code, an inline comment, or a string literal — legacy comment-context matches receive a ×1.5 score weight. The ⚡ bolt icon marks clustered matches: three or more patterns within a 10-line window, each receiving an additional ×1.5 density multiplier under the legacy scoring model.
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | web_demos/web_demo_2.6.py | 9 | CODE | |
| LOW | web_demos/web_demo_2.6.py | 12 | CODE | |
| LOW | web_demos/web_demo_2.6.py | 13 | CODE | |
| LOW | web_demos/web_demo_2.6.py | 14 | CODE | |
| LOW | web_demos/minicpm-o_2.6/model_server.py | 12 | CODE | |
| LOW | web_demos/minicpm-o_2.6/model_server.py | 17 | CODE | |
| LOW | web_demos/minicpm-o_2.6/chatbot_web_demo_o2.6.py | 9 | CODE | |
| LOW | web_demos/minicpm-o_2.6/chatbot_web_demo_o2.6.py | 12 | CODE | |
| LOW | web_demos/minicpm-o_2.6/chatbot_web_demo_o2.6.py | 13 | CODE | |
| LOW | web_demos/minicpm-o_2.6/chatbot_web_demo_o2.6.py | 14 | CODE | |
| LOW | finetune/dataset.py | 2 | CODE | |
| LOW | finetune/dataset.py | 5 | CODE | |
| LOW | finetune/dataset.py | 8 | CODE | |
| LOW | finetune/dataset.py | 8 | CODE | |
| LOW | finetune/dataset.py | 9 | CODE | |
| LOW | finetune/dataset.py | 9 | CODE | |
| LOW | finetune/dataset.py | 16 | CODE | |
| LOW | finetune/dataset.py | 16 | CODE | |
| LOW | finetune/trainer.py | 4 | CODE | |
| LOW | finetune/trainer.py | 8 | CODE | |
| LOW | finetune/trainer.py | 9 | CODE | |
| LOW | finetune/finetune.py | 1 | CODE | |
| LOW | finetune/finetune.py | 7 | CODE | |
| LOW | finetune/finetune.py | 7 | CODE | |
| LOW | finetune/finetune.py | 14 | CODE | |
| LOW | finetune/finetune.py | 15 | CODE | |
| LOW | omnilmm/conversation.py | 3 | CODE | |
| LOW | omnilmm/utils.py | 1 | CODE | |
| LOW | omnilmm/model/__init__.py | 1 | CODE | |
| LOW | omnilmm/model/resampler.py | 6 | CODE | |
| LOW | omnilmm/model/resampler.py | 8 | CODE | |
| LOW | omnilmm/model/resampler.py | 9 | CODE | |
| LOW | omnilmm/model/resampler.py | 11 | CODE | |
| LOW | omnilmm/model/resampler.py | 12 | CODE | |
| LOW | omnilmm/model/resampler.py | 12 | CODE | |
| LOW | omnilmm/model/resampler.py | 12 | CODE | |
| LOW | omnilmm/model/resampler.py | 12 | CODE | |
| LOW | omnilmm/model/resampler.py | 12 | CODE | |
| LOW | omnilmm/model/resampler.py | 12 | CODE | |
| LOW | omnilmm/model/resampler.py | 19 | CODE | |
| LOW | omnilmm/model/resampler.py | 20 | CODE | |
| LOW | omnilmm/train/train_utils.py | 1 | CODE | |
| LOW | omnilmm/train/train_utils.py | 2 | CODE | |
| LOW | omnilmm/train/train_utils.py | 4 | CODE | |
| LOW | omnilmm/train/train_utils.py | 6 | CODE | |
| LOW | omnilmm/train/train_utils.py | 12 | CODE | |
| LOW | omnilmm/train/train_utils.py | 13 | CODE | |
| LOW | eval_mm/vlmevalkit/run.py | 10 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/config.py | 1 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/config.py | 2 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/tools.py | 3 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/tools.py | 4 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/__init__.py | 6 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/__init__.py | 7 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/__init__.py | 8 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/__init__.py | 9 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/__init__.py | 10 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/__init__.py | 11 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/__init__.py | 12 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/__init__.py | 2 | CODE | |
| 196 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | chat.py | 223 | CODE | |
| LOW | web_demos/minicpm-o_2.6/model_server.py | 660 | CODE | |
| LOW | web_demos/minicpm-o_2.6/model_server.py | 884 | CODE | |
| LOW | web_demos/minicpm-o_2.6/model_server.py | 144 | CODE | |
| LOW | web_demos/minicpm-o_2.6/model_server.py | 206 | CODE | |
| LOW | web_demos/minicpm-o_2.6/model_server.py | 312 | CODE | |
| LOW | web_demos/minicpm-o_2.6/model_server.py | 385 | CODE | |
| LOW | web_demos/minicpm-o_2.6/model_server.py | 416 | CODE | |
| LOW | web_demos/minicpm-o_2.6/model_server.py | 489 | CODE | |
| LOW | web_demos/minicpm-o_2.6/vad_utils.py | 103 | CODE | |
| LOW | finetune/dataset.py | 232 | CODE | |
| LOW | finetune/dataset.py | 272 | CODE | |
| LOW | finetune/dataset.py | 321 | CODE | |
| LOW | finetune/dataset.py | 438 | CODE | |
| LOW | finetune/trainer.py | 45 | CODE | |
| LOW | omnilmm/conversation.py | 26 | CODE | |
| LOW | omnilmm/conversation.py | 54 | CODE | |
| LOW | omnilmm/conversation.py | 110 | CODE | |
| LOW | omnilmm/model/omnilmm.py | 123 | CODE | |
| LOW | omnilmm/model/omnilmm.py | 184 | CODE | |
| LOW | eval_mm/vlmevalkit/run.py | 150 | CODE | |
| LOW | eval_mm/vlmevalkit/setup.py | 7 | CODE | |
| LOW | eval_mm/vlmevalkit/setup.py | 24 | CODE | |
| LOW | eval_mm/vlmevalkit/setup.py | 57 | CODE | |
| LOW | eval_mm/vlmevalkit/setup.py | 65 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/tools.py | 268 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/tools.py | 327 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/tools.py | 391 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/inference_video.py | 51 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/image_base.py | 105 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/mmmath.py | 51 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/mmmath.py | 207 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/mmmath.py | 284 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/mmmath.py | 318 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/mmmath.py | 320 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/mmlongbench.py | 112 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/mmlongbench.py | 264 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/mmlongbench.py | 297 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/mmlongbench.py | 460 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/videomme.py | 9 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/videomme.py | 60 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/videomme.py | 186 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/videomme.py | 227 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/videomme.py | 81 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/dynamath.py | 37 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/dynamath.py | 63 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/image_vqa.py | 48 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/image_vqa.py | 149 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/image_vqa.py | 537 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/image_vqa.py | 939 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/image_vqa.py | 968 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/image_vqa.py | 1149 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/image_vqa.py | 1205 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/image_vqa.py | 1323 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/image_yorn.py | 26 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/longvideobench.py | 106 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/longvideobench.py | 271 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/slidevqa.py | 95 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/mlvu.py | 210 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/mmbench_video.py | 11 | CODE | |
| 74 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | chat.py | 119 | except Exception as e: | CODE |
| LOW | chat.py | 148 | except Exception as e: | CODE |
| LOW | chat.py | 172 | except Exception as e: | CODE |
| LOW | chat.py | 228 | except Exception as e: | CODE |
| MEDIUM | chat.py | 116 | def chat(self, input): | CODE |
| MEDIUM | chat.py | 145 | def chat(self, input): | CODE |
| MEDIUM | chat.py | 169 | def chat(self, input): | CODE |
| MEDIUM | web_demos/web_demo_2.6.py | 40 | print('Error: running int4 model with bitsandbytes on Mac is not supported right now.') | CODE |
| LOW | web_demos/web_demo_2.6.py | 147 | except Exception as e: | CODE |
| MEDIUM | web_demos/web_demo_2.6.py | 131 | def chat(img, msgs, ctx, params=None, vision_hidden_states=None): | CODE |
| MEDIUM | web_demos/web_demo_2.5.py | 30 | print('Error: running int4 model with bitsandbytes on Mac is not supported right now.') | CODE |
| LOW | web_demos/web_demo_2.5.py | 154 | except Exception as err: | CODE |
| LOW | web_demos/web_demo.py | 162 | except Exception as err: | CODE |
| MEDIUM | web_demos/minicpm-o_2.6/model_server.py | 290 | def clear(self): | CODE |
| MEDIUM | web_demos/minicpm-o_2.6/model_server.py | 385 | def vad_check_audio_bytes(self, audio, image, sr): | CODE |
| LOW | web_demos/minicpm-o_2.6/model_server.py | 308 | except Exception as e: | CODE |
| LOW | web_demos/minicpm-o_2.6/model_server.py | 370 | except Exception as e: | CODE |
| LOW | web_demos/minicpm-o_2.6/model_server.py | 373 | except Exception as e: | CODE |
| LOW | web_demos/minicpm-o_2.6/model_server.py | 411 | except Exception as e: | CODE |
| LOW | web_demos/minicpm-o_2.6/model_server.py | 476 | except Exception as e: | CODE |
| LOW | web_demos/minicpm-o_2.6/model_server.py | 548 | except Exception as e: | CODE |
| LOW | web_demos/minicpm-o_2.6/model_server.py | 552 | except Exception as e: | CODE |
| LOW | web_demos/minicpm-o_2.6/model_server.py | 590 | except Exception as e: | CODE |
| LOW | web_demos/minicpm-o_2.6/model_server.py | 655 | except Exception as e: | CODE |
| LOW | web_demos/minicpm-o_2.6/model_server.py | 723 | except Exception as e: | CODE |
| LOW | web_demos/minicpm-o_2.6/model_server.py | 730 | except Exception as e: | CODE |
| LOW | web_demos/minicpm-o_2.6/model_server.py | 770 | except Exception as e: | CODE |
| LOW | web_demos/minicpm-o_2.6/model_server.py | 773 | except Exception as e: | CODE |
| LOW | web_demos/minicpm-o_2.6/model_server.py | 818 | except Exception as e: | CODE |
| LOW | web_demos/minicpm-o_2.6/model_server.py | 877 | except Exception as e: | CODE |
| LOW | web_demos/minicpm-o_2.6/model_server.py | 925 | except Exception as e: | CODE |
| LOW | web_demos/minicpm-o_2.6/chatbot_web_demo_o2.6.py | 142 | except Exception as e: | CODE |
| MEDIUM | web_demos/minicpm-o_2.6/chatbot_web_demo_o2.6.py | 126 | def chat(img, msgs, ctx, params=None, vision_hidden_states=None): | CODE |
| LOW | web_demos/minicpm-o_2.6/vad_utils.py | 298 | except Exception as e: | CODE |
| LOW | eval_mm/vlmevalkit/run.py | 410 | except Exception as e: | STRING |
| LOW | eval_mm/vlmevalkit/vlmeval/inference_mt.py | 32 | except Exception as e: | CODE |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/mmlongbench.py | 137 | except Exception: | STRING |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/mmlongbench.py | 200 | except Exception as e: | STRING |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/mmlongbench.py | 464 | except Exception as e: | STRING |
| MEDIUM | eval_mm/vlmevalkit/vlmeval/dataset/mmlongbench.py | 187 | def isfloat(num): | CODE |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/video_base.py | 16 | except Exception as e: | CODE |
| MEDIUM | eval_mm/vlmevalkit/vlmeval/dataset/video_base.py | 9 | def __init__(self, | CODE |
| LOW⚡ | eval_mm/vlmevalkit/vlmeval/dataset/vcr.py | 20 | except Exception as e: | CODE |
| LOW⚡ | eval_mm/vlmevalkit/vlmeval/dataset/vcr.py | 26 | except Exception as e: | CODE |
| LOW⚡ | eval_mm/vlmevalkit/vlmeval/dataset/vcr.py | 34 | except Exception as e: | CODE |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/image_mt.py | 79 | except Exception as e: | CODE |
| LOW⚡ | eval_mm/vlmevalkit/vlmeval/dataset/cgbench.py | 409 | except Exception: | CODE |
| LOW⚡ | eval_mm/vlmevalkit/vlmeval/dataset/cgbench.py | 418 | except Exception: | CODE |
| LOW⚡ | eval_mm/vlmevalkit/vlmeval/dataset/cgbench.py | 427 | except Exception: | CODE |
| LOW⚡ | eval_mm/vlmevalkit/vlmeval/dataset/cgbench.py | 734 | except Exception: | CODE |
| LOW⚡ | eval_mm/vlmevalkit/vlmeval/dataset/cgbench.py | 743 | except Exception: | CODE |
| LOW⚡ | eval_mm/vlmevalkit/vlmeval/dataset/cgbench.py | 752 | except Exception: | CODE |
| LOW⚡ | eval_mm/vlmevalkit/vlmeval/dataset/cgbench.py | 1289 | except Exception: | CODE |
| LOW⚡ | eval_mm/vlmevalkit/vlmeval/dataset/cgbench.py | 1298 | except Exception: | CODE |
| LOW⚡ | eval_mm/vlmevalkit/vlmeval/dataset/cgbench.py | 1307 | except Exception: | CODE |
| LOW⚡ | eval_mm/vlmevalkit/vlmeval/dataset/cgbench.py | 1613 | except Exception: | CODE |
| LOW⚡ | eval_mm/vlmevalkit/vlmeval/dataset/cgbench.py | 1622 | except Exception: | CODE |
| LOW⚡ | eval_mm/vlmevalkit/vlmeval/dataset/cgbench.py | 1631 | except Exception: | CODE |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/mvbench.py | 168 | except Exception as e: | STRING |
| MEDIUM | eval_mm/vlmevalkit/vlmeval/dataset/mvbench.py | 169 | print(f"Error moving {item_path} to {target_path}: {e}") | STRING |
| 34 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH | eval_mm/vlmevalkit/vlmeval/__init__.py | 3 | except ImportError: | CODE |
| HIGH | eval_mm/vlmevalkit/vlmeval/dataset/mmmath.py | 171 | except: | CODE |
| HIGH | eval_mm/vlmevalkit/vlmeval/dataset/mmmath.py | 178 | except: | CODE |
| HIGH | eval_mm/vlmevalkit/vlmeval/dataset/mmmath.py | 186 | except: | CODE |
| HIGH | eval_mm/vlmevalkit/vlmeval/dataset/miabench.py | 94 | except: | CODE |
| HIGH | eval_mm/vlmevalkit/vlmeval/dataset/utils/mathvista.py | 95 | except ValueError: | CODE |
| HIGH | eval_mm/vlmevalkit/vlmeval/dataset/utils/qspatial.py | 84 | except ValueError: | CODE |
| HIGH | …l_mm/vlmevalkit/vlmeval/dataset/utils/olympiadbench.py | 217 | except: | CODE |
| HIGH | …l_mm/vlmevalkit/vlmeval/dataset/utils/olympiadbench.py | 225 | except: | CODE |
| HIGH | …l_mm/vlmevalkit/vlmeval/dataset/utils/olympiadbench.py | 233 | except: | CODE |
| HIGH | eval_mm/vlmevalkit/vlmeval/dataset/utils/mathv.py | 25 | except: | CODE |
| HIGH | eval_mm/vlmevalkit/vlmeval/dataset/utils/mathv.py | 34 | except: | CODE |
| HIGH | eval_mm/vlmevalkit/vlmeval/dataset/utils/mathv.py | 115 | except ValueError: | CODE |
| HIGH | eval_mm/vlmevalkit/vlmeval/dataset/utils/mvbench.py | 96 | except: | CODE |
| HIGH | …dataset/utils/ccocr_evaluator/doc_parsing_evaluator.py | 191 | except: | CODE |
| HIGH | …dataset/utils/ccocr_evaluator/doc_parsing_evaluator.py | 220 | except: | CODE |
| HIGH | eval_mm/vlmevalkit/vlmeval/smp/misc.py | 136 | except ImportError: | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | chat.py | 62 | def expand_question_into_multimodal(question_text, image_token_len, im_st_token, im_ed_token, im_patch_token): | CODE |
| LOW | chat.py | 71 | def wrap_question_for_omni_lmm(question, image_token_len, tokenizer): | CODE |
| LOW | web_demos/web_demo_2.6.py | 314 | def fewshot_add_demonstration(_image, _user_message, _assistant_message, _chat_bot, _app_cfg): | CODE |
| LOW | web_demos/web_demo_2.6.py | 385 | def regenerate_button_clicked(_question, _image, _user_message, _assistant_message, _chat_bot, _app_cfg, params_form): | CODE |
| LOW | web_demos/web_demo_2.5.py | 209 | def regenerate_button_clicked(_question, _chat_bot, _app_cfg, params_form, num_beams, repetition_penalty, repetition_pen | CODE |
| LOW | web_demos/web_demo.py | 217 | def regenerate_button_clicked(_question, _chat_bot, _app_cfg, params_form, num_beams, repetition_penalty, repetition_pen | CODE |
| LOW | web_demos/minicpm-o_2.6/model_server.py | 593 | def update_customized_options(self, uid, options): | CODE |
| LOW | web_demos/minicpm-o_2.6/chatbot_web_demo_o2.6.py | 309 | def fewshot_add_demonstration(_image, _user_message, _assistant_message, _chat_bot, _app_cfg): | CODE |
| LOW | web_demos/minicpm-o_2.6/chatbot_web_demo_o2.6.py | 380 | def regenerate_button_clicked(_question, _image, _user_message, _assistant_message, _chat_bot, _app_cfg, params_form): | CODE |
| LOW | finetune/dataset.py | 200 | def conversation_to_ids_minicpm(conversation, tokenizer): | CODE |
| LOW | finetune/finetune.py | 78 | def safe_save_model_for_hf_trainer(trainer, output_dir: str, bias="none"): | CODE |
| LOW | finetune/finetune.py | 84 | def make_supervised_data_module( | CODE |
| LOW | omnilmm/model/omnilmm.py | 75 | def initialize_vision_modules(self, vision_tower, no_randaug, num_query, image_size, tune_clip=False): | CODE |
| LOW | omnilmm/model/omnilmm.py | 350 | def prepare_inputs_for_generation( | CODE |
| LOW | omnilmm/model/omnilmm.py | 400 | def initialize_vision_tokenizer(self, mm_use_im_start_end, tokenizer, device, | CODE |
| LOW | omnilmm/model/utils.py | 483 | def is_dist_avail_and_initialized(): | CODE |
| LOW | eval_mm/vlmevalkit/run.py | 30 | def build_dataset_from_config(cfg, dataset_name): | CODE |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/mmmath.py | 47 | def trans_greater_sign_to_interval(self, expr:str): | CODE |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/longvideobench.py | 26 | def insert_subtitles_into_frames( | CODE |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/cmmmu.py | 10 | def get_multi_choice_prediction(response, all_choices, index2ans): | CODE |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/cmmmu.py | 108 | def get_fill_blank_prediction(response, answer): | CODE |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/mvbench.py | 281 | def load_into_video_and_process(self, line): | STRING |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/utils/tempcompass.py | 69 | def eval_rule_caption_matching(line): | STRING |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/utils/tempcompass.py | 203 | def evaluate_tempcompass_captioning(model, line): | STRING |
| LOW | …l_mm/vlmevalkit/vlmeval/dataset/utils/olympiadbench.py | 28 | def get_single_answer_type_text(answer_type, is_chinese): | CODE |
| LOW | …l_mm/vlmevalkit/vlmeval/dataset/utils/olympiadbench.py | 531 | def calculate_merged_accuracy(reference_dir, text_only): | CODE |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/utils/wemath.py | 44 | def evaluate_process_steps_data(df, steps): | CODE |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/utils/wemath.py | 61 | def evaluate_calculate_metrics(merged_2steps, merged_3steps): | CODE |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/utils/wemath.py | 128 | def evaluate_compute_final_scores(metrics, total_count): | CODE |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/utils/wemath.py | 159 | def evaluate_update_main_results_df(main_results_df, total_counts, rates): | CODE |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/utils/wemath.py | 237 | def load_knowledge_structure_nodes(filepath): | CODE |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/utils/wemath.py | 267 | def accuracy_process_steps_data(df, steps): | CODE |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/utils/wemath.py | 284 | def accuracy_update_main_results_df(nodes, main_results_df, concatenated_data, merged_2steps, merged_3steps): | CODE |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/utils/mlvu.py | 98 | def check_ans_with_model_summary(pred, gt, model, item, dataset_name='MLVU_OpenEnded'): | STRING |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/utils/mlvu.py | 132 | def check_ans_with_model_sub_scene(pred, gt, model, item, dataset_name='MLVU_OpenEnded'): | STRING |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/utils/cgbench.py | 87 | def get_dimention_rating_open_ended(data_path): | CODE |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/utils/cgbench.py | 128 | def get_dimention_rating_mcq_grouding(data_path): | CODE |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/utils/cgbench.py | 201 | def calculate_acc_at_iou_threshold(scores, threshold): | CODE |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/utils/cgbench.py | 243 | def calculate_rec_at_iou_threshold(scores, threshold): | CODE |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/utils/cgbench.py | 286 | def sample_frames_clue_average(clues_time_intervals, frame_num, fps): | CODE |
| LOW | …vlmeval/dataset/utils/ccocr_evaluator/kie_evaluator.py | 275 | def normalize_values_of_nested_dict(d, normalize_func): | CODE |
| LOW | …vlmeval/dataset/utils/ccocr_evaluator/kie_evaluator.py | 335 | def remove_unnecessary_spaces(text): | CODE |
| LOW | …vlmeval/dataset/utils/ccocr_evaluator/ocr_evaluator.py | 21 | def text_normalize_and_tokenize(text, is_keep_blank=True, is_lower=True, is_alphanum_only=False): | CODE |
| LOW | eval_mm/vlmevalkit/vlmeval/vlm/base.py | 169 | def message_to_promptvideo_withrole(self, message, dataset=None): | CODE |
| LOW⚡ | eval_mm/vlmevalkit/vlmeval/api/base.py | 197 | def preprocess_message_with_role(self, message): | CODE |
| LOW | eval_mm/vqaeval/models/MiniCPM/minicpmv.py | 56 | def generate_with_interleaved(self, images, questions, datasetname): | CODE |
| LOW | eval_mm/vqaeval/models/MiniCPM/minicpmv.py | 147 | def generate_with_interleaved(self, images, questions, datasetname): | CODE |
| LOW | eval_mm/vqaeval/models/MiniCPM/minicpmv.py | 256 | def generate_with_interleaved(self, images, questions, datasetname): | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | chat.py | 131 | CODE | |
| LOW | omnilmm/model/omnilmm.py | 121 | CODE | |
| LOW | omnilmm/model/omnilmm.py | 184 | CODE | |
| LOW | omnilmm/model/omnilmm.py | 283 | CODE | |
| LOW | omnilmm/model/utils.py | 115 | CODE | |
| LOW | omnilmm/model/utils.py | 139 | CODE | |
| LOW | omnilmm/model/utils.py | 150 | CODE | |
| LOW | omnilmm/model/utils.py | 160 | CODE | |
| LOW | omnilmm/model/utils.py | 182 | CODE | |
| LOW | omnilmm/model/utils.py | 195 | CODE | |
| LOW | omnilmm/model/utils.py | 205 | CODE | |
| LOW | omnilmm/model/utils.py | 235 | CODE | |
| LOW | omnilmm/model/utils.py | 246 | CODE | |
| LOW | omnilmm/model/utils.py | 257 | CODE | |
| LOW | omnilmm/model/utils.py | 265 | CODE | |
| LOW | omnilmm/model/utils.py | 273 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/inference_video.py | 48 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/mmlongbench.py | 420 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/vcr.py | 171 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/slidevqa.py | 53 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/cmmmu.py | 321 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/dude.py | 34 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/utils/mathvista.py | 164 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/utils/yorn.py | 42 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/utils/yorn.py | 88 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/utils/yorn.py | 140 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/utils/yorn.py | 182 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/utils/yorn.py | 204 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/utils/logicvista.py | 150 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/utils/mlvu.py | 107 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/utils/mlvu.py | 142 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/utils/mlvu.py | 165 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/utils/mathv.py | 171 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/utils/cgbench.py | 504 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/utils/cgbench.py | 530 | CODE | |
| LOW | …vlmeval/dataset/utils/ccocr_evaluator/kie_evaluator.py | 332 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/vlm/minicpm_v.py | 727 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/smp/misc.py | 161 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/smp/vlm.py | 93 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/smp/file.py | 175 | CODE | |
| LOW | eval_mm/vlmevalkit/vlmeval/api/gpt.py | 36 | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/mmmath.py | 98 | # Check if two < or > in expression | COMMENT |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/mmmath.py | 118 | # Check if elements in both lists can be paired and are equal | COMMENT |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/mmmath.py | 192 | # Check if two numerical values are equal within an allowed error range | COMMENT |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/mmmath.py | 208 | # Check if two expressions are mathematically equivalent | COMMENT |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/mmmath.py | 258 | # Check if two equations are mathematically equivalent | COMMENT |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/mmmath.py | 285 | # Check if two intervals are mathematically equivalent | COMMENT |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/videomme.py | 96 | # Check if the member is a file (not a directory) | STRING |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/videomme.py | 114 | # Check if the member is a file (not a directory) | STRING |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/image_mcq.py | 641 | # Check if dataset is cached and has integrity | COMMENT |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/utils/tempcompass.py | 154 | # Check if the chatgpt answer is the ground-truth answer | STRING |
| LOW⚡ | …l_mm/vlmevalkit/vlmeval/dataset/utils/olympiadbench.py | 475 | # Check if the expression is a power expression | COMMENT |
| LOW⚡ | …l_mm/vlmevalkit/vlmeval/dataset/utils/olympiadbench.py | 480 | # Check if the base and the exponent are numbers | COMMENT |
| LOW⚡ | …l_mm/vlmevalkit/vlmeval/dataset/utils/olympiadbench.py | 485 | # Check if the exponent is greater than the threshold | COMMENT |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/utils/cgbench.py | 320 | # Check if there is an overlap | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | README.md | 1633 | # Step 1: Streaming prefill | COMMENT |
| LOW | README.md | 1641 | # Step 2: Streaming generate | COMMENT |
| LOW⚡ | README.md | 1776 | # Step 1: Prefill system prompt | COMMENT |
| LOW⚡ | README.md | 1780 | # Step 2: Prefill omni chunks (is_last_chunk=True only for the last audio chunk) | COMMENT |
| LOW⚡ | README.md | 1789 | # Step 3: Generate response | COMMENT |
| LOW | README_zh.md | 1614 | # Step 1: Streaming prefill | COMMENT |
| LOW | README_zh.md | 1622 | # Step 2: Streaming generate | COMMENT |
| LOW⚡ | README_zh.md | 1757 | # Step 1: Prefill system prompt | COMMENT |
| LOW⚡ | README_zh.md | 1761 | # Step 2: Prefill omni chunks (is_last_chunk=True only for the last audio chunk) | COMMENT |
| LOW⚡ | README_zh.md | 1770 | # Step 3: Generate response | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH | eval_mm/vlmevalkit/vlmeval/dataset/utils/mathvista.py | 0 | please read the following example. then extract the answer from the model response and type it at the end of the prompt. | STRING |
| HIGH | eval_mm/vlmevalkit/vlmeval/dataset/utils/qspatial.py | 0 | please read the following example. then extract the answer from the model response and type it at the end of the prompt. | STRING |
| HIGH | eval_mm/vlmevalkit/vlmeval/dataset/utils/mathv.py | 0 | please read the following example. then extract the answer from the model response and type it at the end of the prompt. | STRING |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | …-o_2.6/web_server/src/components/ModelConfig/index.vue | 181 | base64Str: '' | COMMENT |
| LOW | …2.6/web_server/src/views/home/components/VideoCall.vue | 601 | }); | COMMENT |
| LOW | …eb_server/src/views/home/components/VoiceCall_0105.vue | 501 | // '截取前长度:', | COMMENT |
| LOW | …eb_server/src/views/home/components/VoiceCall_0105.vue | 521 | item().finally(() => playAudio()); | COMMENT |
| LOW | …eb_server/src/views/home/components/VideoCall_0105.vue | 561 | // skipDisabled.value = true; | COMMENT |
| LOW | …eb_server/src/views/home/components/VideoCall_0105.vue | 601 | // vadStartTime.value = +new Date(); | COMMENT |
| LOW | …2.6/web_server/src/views/home/components/VoiceCall.vue | 501 | if (isCalling.value && !isReturnError.value) { | COMMENT |
| LOW | …2.6/web_server/src/views/home/components/VoiceCall.vue | 541 | // taskQueue.value.map(item => item.time) | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| MEDIUM | web_demos/minicpm-o_2.6/model_server.py | 863 | # Define the log file path | COMMENT |
| MEDIUM | web_demos/minicpm-o_2.6/vad_utils.py | 108 | """This method is used for splitting long audios into speech chunks using silero VAD. | STRING |
| MEDIUM | eval_mm/vlmevalkit/run.py | 399 | # Create the symbolic links for the prediction files | STRING |
| MEDIUM | eval_mm/vlmevalkit/vlmeval/dataset/utils/mlvu.py | 77 | # Define the keys to locate in the text | STRING |
| MEDIUM | eval_mm/vlmevalkit/vlmeval/dataset/utils/mlvu.py | 111 | # Define the keys to locate in the text | STRING |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| MEDIUM | eval_mm/vlmevalkit/vlmeval/api/gpt.py | 185 | # 'may exceed the context window with some additional meta symbols. ' | COMMENT |
| MEDIUM | eval_mm/vlmevalkit/vlmeval/api/gpt.py | 188 | # return 0, self.fail_msg + 'Input string longer than context window. ', 'Length Exceeded. ' | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH | docs/api.md | 17 | export API_KEY="your-api-key" | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | finetune/dataset.py | 19 | logger = logging.getLogger(__name__) | CODE |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/__init__.py | 235 | __all__ = [ | CODE |
| LOW | eval_mm/vlmevalkit/vlmeval/dataset/utils/__init__.py | 6 | __all__ = [ | CODE |
| LOW | eval_mm/vlmevalkit/vlmeval/utils/__init__.py | 5 | __all__ = [ | CODE |
| LOW | eval_mm/vlmevalkit/vlmeval/api/__init__.py | 3 | __all__ = [ | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| MEDIUM | eval_mm/vlmevalkit/vlmeval/dataset/image_vqa.py | 1080 | # Once you get the permission, you can use the helper code here to download and extract necessary images: | COMMENT |
| MEDIUM | eval_mm/vlmevalkit/vlmeval/dataset/utils/mathverse.py | 29 | Model response: 'As it stands, I cannot provide the correct option letter because there isn't enough information to solv | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH | …l_mm/vlmevalkit/vlmeval/dataset/utils/tablevqabench.py | 23 | 'If you have multiple answers, please separate them with || marks. Example: Apple||Banana||Tomato\n\n' | CODE |