Scan timing: clone 9.94s · analysis 11.53s · 13.1 MB · GitHub API rate-limit (preflight)
https://github.com/hiyouga/LlamaFactory
· scanned 2026-06-05 08:17 UTC (5 days, 21 hours ago)
· 10 languages
446 raw signals (216 security + 230 graph) 53rd percentile · Python · medium (20-100K LoC) System graph score 82 (lower by 16)
Last scanned 5 days, 21 hours ago · v2 · 168 actionable findings from 2 signal sources. 163 repeated signals grouped for readability. Security checks, system graph analysis, and verified AI-agent feedback are merged into one review queue.
| Component | Sub-score | Weight | Contribution |
|---|---|---|---|
structure_score |
85.0 | 0.15 | 12.75 |
security_score |
34.4 | 0.25 | 8.60 |
testing_score |
94.0 | 0.20 | 18.80 |
documentation_score |
73.0 | 0.15 | 10.95 |
practices_score |
72.0 | 0.15 | 10.80 |
code_quality |
41.6 | 0.10 | 4.16 |
| Overall | 1.00 | 66.1 |
All 1925 nodes from the latest scan, grouped by kind. Each node is a unit the engine identified (file, function, endpoint, table…). Most users won't need this view — it's primarily for debugging the engine's graph extraction or for AI agents that want to enumerate the project structure.
| Label | Layer | Status | Path |
|---|---|---|---|
pytest_configure |
software | healthy | tests/conftest.py:35 |
_handle_runs_on |
software | healthy | tests/conftest.py:51 |
_handle_slow_tests |
software | healthy | tests/conftest.py:66 |
_get_visible_devices_env |
software | healthy | tests/conftest.py:75 |
_handle_device_visibility |
software | healthy | tests/conftest.py:85 |
pytest_collection_modifyitems |
software | healthy | tests/conftest.py:109 |
_cleanup_distributed_state |
software | healthy | tests/conftest.py:123 |
_manage_distributed_env |
software | healthy | tests/conftest.py:131 |
fix_valuehead_cpu_loading |
software | healthy | tests/conftest.py:166 |
bypass_mistral_regex_check |
software | healthy | tests/conftest.py:172 |
main |
software | healthy | tests/check_license.py:22 |
test_eval_template_en |
software | healthy | tests/eval/test_eval_template.py:21 |
test_eval_template_zh |
software | healthy | tests/eval/test_eval_template.py:60 |
__call__ |
software | healthy | tests/train/test_sft_trainer.py:52 |
test_shuffle |
software | healthy | tests/train/test_sft_trainer.py:63 |
test_lora_train_qv_modules |
software | healthy | tests/model/test_lora.py:58 |
test_lora_train_all_modules |
software | healthy | tests/model/test_lora.py:64 |
test_lora_train_extra_modules |
software | healthy | tests/model/test_lora.py:70 |
test_lora_train_old_adapters |
software | healthy | tests/model/test_lora.py:76 |
test_lora_train_new_adapters |
software | healthy | tests/model/test_lora.py:82 |
test_lora_train_valuehead |
software | healthy | tests/model/test_lora.py:91 |
test_lora_inference |
software | healthy | tests/model/test_lora.py:100 |
test_base |
software | healthy | tests/model/test_base.py:33 |
test_valuehead |
software | healthy | tests/model/test_base.py:40 |
test_full_train |
software | healthy | tests/model/test_full.py:46 |
test_full_inference |
software | healthy | tests/model/test_full.py:53 |
test_freeze_train_all_modules |
software | healthy | tests/model/test_freeze.py:46 |
test_freeze_train_extra_modules |
software | healthy | tests/model/test_freeze.py:57 |
test_freeze_inference |
software | healthy | tests/model/test_freeze.py:68 |
test_pissa_train |
software | healthy | tests/model/test_pissa.py:53 |
test_pissa_inference |
software | healthy | tests/model/test_pissa.py:60 |
test_get_seqlens_in_batch |
software | healthy | tests/model/model_utils/test_packing.py:37 |
test_get_unpad_data |
software | healthy | tests/model/model_utils/test_packing.py:63 |
__init__ |
software | healthy | tests/model/model_utils/test_embedding.py:44 |
convert_tokens_to_ids |
software | healthy | tests/model/model_utils/test_embedding.py:34 |
__call__ |
software | healthy | tests/model/model_utils/test_embedding.py:37 |
get_input_embeddings |
software | healthy | tests/model/model_utils/test_embedding.py:47 |
test_resolve_new_token_ids_returns_none_without_config |
software | healthy | tests/model/model_utils/test_embedding.py:51 |
test_resolve_new_token_ids_filters_invalid_and_dedups |
software | healthy | tests/model/model_utils/test_embedding.py:57 |
test_existing_embeddings_excludes_new_token_ids |
software | healthy | tests/model/model_utils/test_embedding.py:67 |
test_noisy_mean_initialization_with_token_ids_targets_exact… |
software | healthy | tests/model/model_utils/test_embedding.py:80 |
test_noisy_mean_initialization_tail_fallback |
software | healthy | tests/model/model_utils/test_embedding.py:103 |
test_description_init_excludes_new_token_ids_from_average |
software | healthy | tests/model/model_utils/test_embedding.py:118 |
is_torch_sdpa_available |
software | healthy | tests/model/model_utils/test_attention.py:26 |
test_attention |
software | healthy | tests/model/model_utils/test_attention.py:43 |
test_vanilla_checkpointing |
software | healthy | tests/model/model_utils/test_checkpointing.py:43 |
test_unsloth_gradient_checkpointing |
software | healthy | tests/model/model_utils/test_checkpointing.py:49 |
test_upcast_layernorm |
software | healthy | tests/model/model_utils/test_checkpointing.py:55 |
test_upcast_lmhead_output |
software | healthy | tests/model/model_utils/test_checkpointing.py:62 |
test_add_tokens |
software | healthy | tests/model/model_utils/test_add_tokens.py:29 |
Showing first 50 of this kind. Full payload available via the JSON button at the top of the page.
| Label | Layer | Status | Path |
|---|---|---|---|
README_zh.md |
software | healthy | README_zh.md |
README.md |
software | healthy | README.md |
.pre-commit-config.yaml |
software | healthy | .pre-commit-config.yaml |
pyproject.toml |
software | healthy | pyproject.toml |
Makefile |
software | healthy | Makefile |
Dockerfile |
software | healthy | docker/docker-rocm/Dockerfile |
docker-compose.yml |
software | healthy | docker/docker-rocm/docker-compose.yml |
README.md |
software | healthy | docker/docker-cuda/README.md |
Dockerfile |
software | healthy | docker/docker-cuda/Dockerfile |
docker-compose.yml |
software | healthy | docker/docker-cuda/docker-compose.yml |
Dockerfile |
software | healthy | docker/docker-npu/Dockerfile |
docker-compose.yml |
software | healthy | docker/docker-npu/docker-compose.yml |
conftest.py |
software | healthy | tests/conftest.py |
check_license.py |
software | healthy | tests/check_license.py |
test_eval_template.py |
software | healthy | tests/eval/test_eval_template.py |
test_sft_trainer.py |
software | healthy | tests/train/test_sft_trainer.py |
test_lora.py |
software | healthy | tests/model/test_lora.py |
test_base.py |
software | healthy | tests/model/test_base.py |
test_full.py |
software | healthy | tests/model/test_full.py |
test_freeze.py |
software | healthy | tests/model/test_freeze.py |
test_pissa.py |
software | healthy | tests/model/test_pissa.py |
test_packing.py |
software | healthy | tests/model/model_utils/test_packing.py |
test_embedding.py |
software | healthy | tests/model/model_utils/test_embedding.py |
test_attention.py |
software | healthy | tests/model/model_utils/test_attention.py |
test_checkpointing.py |
software | healthy | tests/model/model_utils/test_checkpointing.py |
test_add_tokens.py |
software | healthy | tests/model/model_utils/test_add_tokens.py |
test_visual.py |
software | healthy | tests/model/model_utils/test_visual.py |
test_misc.py |
software | healthy | tests/model/model_utils/test_misc.py |
test_sglang.py |
software | healthy | tests/e2e/test_sglang.py |
test_train.py |
software | healthy | tests/e2e/test_train.py |
test_chat.py |
software | healthy | tests/e2e/test_chat.py |
conf.py |
software | warning | docs/conf.py |
Makefile |
software | healthy | docs/Makefile |
requirements.txt |
software | healthy | docs/requirements.txt |
conf.py |
software | warning | docs/zh/conf.py |
getting-started.md |
software | healthy | docs/zh/getting-started.md |
installation.md |
software | healthy | docs/zh/installation.md |
llamaboard-web-ui.md |
software | healthy | docs/zh/llamaboard-web-ui.md |
model-argument.md |
software | healthy | docs/zh/hyperparameters/model-argument.md |
sample-argument.md |
software | healthy | docs/zh/hyperparameters/sample-argument.md |
data-argument.md |
software | healthy | docs/zh/hyperparameters/data-argument.md |
training-argument.md |
software | healthy | docs/zh/hyperparameters/training-argument.md |
data-processing.md |
software | healthy | docs/zh/data-preparation/data-processing.md |
deploy.md |
software | healthy | docs/zh/inference/deploy.md |
data-engine.md |
software | healthy | docs/zh/dev-guide/core/data-engine.md |
model-engine.md |
software | healthy | docs/zh/dev-guide/core/model-engine.md |
trainer.md |
software | healthy | docs/zh/dev-guide/core/trainer.md |
data-plugins.md |
software | healthy | docs/zh/dev-guide/plugins/data-plugins.md |
rendering.md |
software | healthy | docs/zh/dev-guide/plugins/model-plugins/rendering.md |
kernels.md |
software | healthy | docs/zh/dev-guide/plugins/model-plugins/kernels.md |
Showing first 50 of this kind. Full payload available via the JSON button at the top of the page.
| Label | Layer | Status | Path |
|---|---|---|---|
DataCollatorWithVerbose |
software | healthy | tests/train/test_sft_trainer.py:49 |
_StubTokenizer |
software | healthy | tests/model/model_utils/test_embedding.py:25 |
_StubModel |
software | healthy | tests/model/model_utils/test_embedding.py:41 |
DummyDataset |
software | healthy | scripts/bench_qwen.py:32 |
MultiModalDataCollatorForSeq2Seq |
software | healthy | scripts/bench_qwen.py:79 |
PairwiseDataCollatorWithPadding |
software | healthy | scripts/stat_utils/cal_ppl.py:32 |
_RestartableDataProvider |
software | healthy | tests_v1/core/utils/test_batching.py:38 |
Holder |
software | healthy | tests_v1/plugins/trainer_plugins/distributed/test_fsdp2_wei… |
FakeFusedExpertsModel |
software | healthy | tests_v1/plugins/trainer_plugins/distributed/test_fsdp2_wei… |
FakeLegacyExpert |
software | healthy | tests_v1/plugins/trainer_plugins/distributed/test_fsdp2_wei… |
FakeLegacyExpertsModel |
software | healthy | tests_v1/plugins/trainer_plugins/distributed/test_fsdp2_wei… |
Engine |
software | healthy | src/llamafactory/webui/engine.py:28 |
Manager |
software | healthy | src/llamafactory/webui/manager.py:23 |
WebChatModel |
software | healthy | src/llamafactory/webui/chatter.py:80 |
Runner |
software | healthy | src/llamafactory/webui/runner.py:54 |
RMTrainer |
software | healthy | src/llamafactory/v1/trainers/rm_trainer.py:70 |
SFTTrainer |
software | healthy | src/llamafactory/v1/trainers/sft_trainer.py:24 |
PluginConfig |
software | healthy | src/llamafactory/v1/config/arg_utils.py:23 |
ModelClass |
software | healthy | src/llamafactory/v1/config/arg_utils.py:39 |
SampleBackend |
software | healthy | src/llamafactory/v1/config/arg_utils.py:48 |
BatchingStrategy |
software | healthy | src/llamafactory/v1/config/arg_utils.py:54 |
TrainingArguments |
software | healthy | src/llamafactory/v1/config/training_args.py:23 |
SampleArguments |
software | healthy | src/llamafactory/v1/config/sample_args.py:22 |
DataArguments |
software | healthy | src/llamafactory/v1/config/data_args.py:20 |
ModelArguments |
software | healthy | src/llamafactory/v1/config/model_args.py:23 |
AttentionFunction |
software | healthy | src/llamafactory/v1/utils/types.py:58 |
DatasetInfo |
software | healthy | src/llamafactory/v1/utils/types.py:64 |
DistributedConfig |
software | healthy | src/llamafactory/v1/utils/types.py:81 |
Content |
software | healthy | src/llamafactory/v1/utils/types.py:94 |
Message |
software | healthy | src/llamafactory/v1/utils/types.py:101 |
SFTSample |
software | healthy | src/llamafactory/v1/utils/types.py:110 |
DPOSample |
software | healthy | src/llamafactory/v1/utils/types.py:121 |
ToolCall |
software | healthy | src/llamafactory/v1/utils/types.py:137 |
ModelInput |
software | healthy | src/llamafactory/v1/utils/types.py:144 |
BatchInput |
software | healthy | src/llamafactory/v1/utils/types.py:159 |
BatchInfo |
software | healthy | src/llamafactory/v1/utils/types.py:174 |
ModelOutput |
software | healthy | src/llamafactory/v1/utils/types.py:183 |
StatefulBuffer |
software | healthy | src/llamafactory/v1/utils/objects.py:21 |
BasePlugin |
software | healthy | src/llamafactory/v1/utils/plugin.py:26 |
PrintPlugin |
software | healthy | src/llamafactory/v1/utils/plugin.py:94 |
_Logger |
software | healthy | src/llamafactory/v1/utils/logging.py:31 |
DtypeRegistry |
software | healthy | src/llamafactory/v1/utils/dtype.py:26 |
DtypeInterface |
software | healthy | src/llamafactory/v1/utils/dtype.py:32 |
TrainerState |
software | healthy | src/llamafactory/v1/utils/callbacks/trainer_callback.py:26 |
TrainerCallback |
software | healthy | src/llamafactory/v1/utils/callbacks/trainer_callback.py:48 |
CallbackHandler |
software | healthy | src/llamafactory/v1/utils/callbacks/trainer_callback.py:98 |
LoggingCallback |
software | healthy | src/llamafactory/v1/utils/callbacks/logging_callback.py:32 |
SyncSampler |
software | healthy | src/llamafactory/v1/samplers/cli_sampler.py:28 |
ModelEngine |
software | healthy | src/llamafactory/v1/core/model_engine.py:47 |
BaseSampler |
software | healthy | src/llamafactory/v1/core/base_sampler.py:23 |
Showing first 50 of this kind. Full payload available via the JSON button at the top of the page.
| Label | Layer | Status | Path |
|---|---|---|---|
docker |
software | healthy | docker |
docker-rocm |
software | healthy | docker/docker-rocm |
docker-cuda |
software | healthy | docker/docker-cuda |
docker-npu |
software | healthy | docker/docker-npu |
tests |
software | healthy | tests |
eval |
software | healthy | tests/eval |
train |
software | healthy | tests/train |
model |
software | healthy | tests/model |
model_utils |
software | healthy | tests/model/model_utils |
e2e |
software | healthy | tests/e2e |
docs |
software | healthy | docs |
zh |
software | healthy | docs/zh |
hyperparameters |
software | healthy | docs/zh/hyperparameters |
data-preparation |
software | healthy | docs/zh/data-preparation |
inference |
software | healthy | docs/zh/inference |
dev-guide |
software | healthy | docs/zh/dev-guide |
core |
software | healthy | docs/zh/dev-guide/core |
plugins |
software | healthy | docs/zh/dev-guide/plugins |
model-plugins |
software | healthy | docs/zh/dev-guide/plugins/model-plugins |
advanced |
software | healthy | docs/zh/advanced |
custom-kernels |
software | healthy | docs/zh/advanced/custom-kernels |
distributed |
software | healthy | docs/zh/advanced/distributed |
lora-and-quantization |
software | healthy | docs/zh/advanced/lora-and-quantization |
training |
software | healthy | docs/zh/training |
_static |
software | healthy | docs/_static |
js |
software | healthy | docs/_static/js |
css |
software | healthy | docs/_static/css |
en |
software | healthy | docs/en |
hyperparameters |
software | healthy | docs/en/hyperparameters |
data-preparation |
software | healthy | docs/en/data-preparation |
inference |
software | healthy | docs/en/inference |
dev-guide |
software | healthy | docs/en/dev-guide |
core |
software | healthy | docs/en/dev-guide/core |
plugins |
software | healthy | docs/en/dev-guide/plugins |
model-plugins |
software | healthy | docs/en/dev-guide/plugins/model-plugins |
advanced |
software | healthy | docs/en/advanced |
custom-kernels |
software | healthy | docs/en/advanced/custom-kernels |
distributed |
software | healthy | docs/en/advanced/distributed |
lora-and-quantization |
software | healthy | docs/en/advanced/lora-and-quantization |
training |
software | healthy | docs/en/training |
scripts |
software | healthy | scripts |
convert_ckpt |
software | healthy | scripts/convert_ckpt |
api_example |
software | healthy | scripts/api_example |
stat_utils |
software | healthy | scripts/stat_utils |
examples |
software | healthy | examples |
megatron |
software | healthy | examples/megatron |
accelerate |
software | healthy | examples/accelerate |
v1 |
software | healthy | examples/v1 |
train_qlora |
software | healthy | examples/v1/train_qlora |
train_freeze |
software | healthy | examples/v1/train_freeze |
Showing first 50 of this kind. Full payload available via the JSON button at the top of the page.
| Label | Layer | Status | Path |
|---|---|---|---|
tests |
cicd | healthy | .github/workflows/tests_cuda.yml |
build |
cicd | healthy | .github/workflows/docker.yml |
tests |
cicd | healthy | .github/workflows/tests_npu.yml |
label_issue |
cicd | healthy | .github/workflows/label_issue.yml |
build |
cicd | healthy | .github/workflows/docs.yml |
deploy |
cicd | healthy | .github/workflows/docs.yml |
tests |
cicd | healthy | .github/workflows/tests.yml |
publish |
cicd | healthy | .github/workflows/publish.yml |
| Label | Layer | Status | Path |
|---|---|---|---|
image::docker/docker-rocm/Dockerfile |
hardware | healthy | docker/docker-rocm/Dockerfile |
image::docker/docker-cuda/Dockerfile |
hardware | healthy | docker/docker-cuda/Dockerfile |
image::docker/docker-npu/Dockerfile |
hardware | healthy | docker/docker-npu/Dockerfile |
llamafactory |
hardware | healthy | docker/docker-rocm/docker-compose.yml |
llamafactory |
hardware | healthy | docker/docker-cuda/docker-compose.yml |
llamafactory-a2 |
hardware | healthy | docker/docker-npu/docker-compose.yml |
llamafactory-a3 |
hardware | healthy | docker/docker-npu/docker-compose.yml |
| Label | Layer | Status | Path |
|---|---|---|---|
gha::tests_cuda |
cicd | healthy | .github/workflows/tests_cuda.yml |
gha::docker |
cicd | healthy | .github/workflows/docker.yml |
gha::tests_npu |
cicd | healthy | .github/workflows/tests_npu.yml |
gha::label_issue |
cicd | healthy | .github/workflows/label_issue.yml |
gha::docs |
cicd | healthy | .github/workflows/docs.yml |
gha::tests |
cicd | healthy | .github/workflows/tests.yml |
gha::publish |
cicd | healthy | .github/workflows/publish.yml |
| Label | Layer | Status | Path |
|---|---|---|---|
port:7860 |
network | healthy | docker/docker-rocm/Dockerfile |
port:8000 |
network | healthy | docker/docker-rocm/Dockerfile |
port:7861 |
network | healthy | docker/docker-npu/docker-compose.yml |
port:8001 |
network | healthy | docker/docker-npu/docker-compose.yml |
| Label | Layer | Status | Path |
|---|---|---|---|
GITHUB_TOKEN |
cicd | healthy | — |
QUAY_ASCEND_TOKEN |
cicd | healthy | — |
HF_TOKEN |
cicd | healthy | — |
DOCKERHUB_TOKEN |
cicd | healthy | — |
| Label | Layer | Status | Path |
|---|---|---|---|
GET /v1/models |
api | healthy | src/llamafactory/api/app.py |
POST /v1/chat/completions |
api | healthy | src/llamafactory/api/app.py |
POST /v1/score/evaluation |
api | healthy | src/llamafactory/api/app.py |
| Label | Layer | Status | Path |
|---|---|---|---|
127.0.0.1 |
network | healthy | docker/docker-rocm/Dockerfile |
192.168.0.1 |
network | healthy | examples/accelerate/fsdp_config_multiple_nodes.yaml |
| Label | Layer | Status | Path |
|---|---|---|---|
repobility-clone-gm9dvgdj |
software | healthy | /tmp/repobility-clone-gm9dvgdj |
| Label | Layer | Status | Path |
|---|---|---|---|
nginx |
network | healthy | .github/workflows/tests_npu.yml |
| Label | Layer | Status | Path |
|---|---|---|---|
gpu (detected) |
hardware | healthy | docker/docker-rocm/Dockerfile |
This page is publicly accessible at:
https://repobility.com/scan/8370691f-4f2c-4b46-ba68-f1d31066d05f/
To check status programmatically (no auth required):
curl -s https://repobility.com/api/v1/public/scan/8370691f-4f2c-4b46-ba68-f1d31066d05f/
Important — please don't re-submit the same URL repeatedly. The submission endpoint is idempotent: re-submitting the same git URL returns this same scan_token, not a new one. To re-scan this repo, sign up free and use the dashboard.