absl-py
accelerate>=0.19.0
cpm_kernels
datasets<=3.6.0,>=2.12.0
einops>=0.5.0
evaluate>=0.3.0
func_timeout
fuzzywuzzy
gradio-client
h5py
httpx
immutabledict
importlib-metadata
jieba
json5
jsonlines
mmengine-lite
nltk>=3.7
numpy<2.0.0,>=1.23.4
openai
opencv-python-headless
orjson
pandas
plotly
prettytable
protobuf
pyext==0.5
python-Levenshtein
rank_bm25==0.2.2
rapidfuzz
requests>=2.31.0
retrying
rich
rouge
rouge_chinese
rouge_score
sacrebleu
scikit_learn>=1.5.0
scipy
seaborn
tabulate
tiktoken
timeout_decorator
tokenizers
torch
tqdm>=4.64.1
transformers
tree-sitter==0.21.3
tree_sitter_languages>=1.10.2
typer
Pillow==11.2.1
janus

[bfcl]
bfcl_eval==2025.6.14
annotated-types==0.7.0
anyio==4.10.0
argcomplete==3.6.2
black==25.1.0
cachetools==5.5.2
certifi==2025.8.3
charset-normalizer==3.4.3
click==8.2.1
cohere==5.13.3
datamodel-code-generator==0.25.7
distro==1.9.0
eval_type_backport==0.2.2
exceptiongroup==1.3.0
fastavro==1.12.0
filelock==3.19.1
fsspec==2024.12.0
genson==1.3.0
google-api-core==2.25.1
google-auth==2.40.3
google-cloud-aiplatform==1.93.1
google-cloud-storage==2.19.0
googleapis-common-protos==1.70.0
h11==0.16.0
hf-xet
httpcore==1.0.9
httpx-sse==0.4.0
huggingface-hub
idna==3.10
inflect==5.6.2
isort==5.13.2
jiter==0.10.0
Jinja2==3.1.6
Markdown-it-py==4.0.0
MarkupSafe==3.0.2
mdurl==0.1.2
mistralai==1.7.0
mypy_extensions==1.1.0
packaging==25.0
parameterized==0.9.0
pathspec==0.12.1
platformdirs==4.3.8
proto-plus==1.26.1
pydantic==2.11.7
pydantic_core==2.33.2
PyASN1==0.6.1
pyasn1_modules==0.4.2
Pygments==2.19.2
python-dateutil==2.9.0.post0
python-dotenv==1.1.1
pytz==2025.2
PyYAML==6.0.2
shellingham==1.5.4
six==1.17.0
sniffio==1.3.1
toml==0.10.2
tomli==2.2.1
typing-inspection==0.4.1
typing_extensions==4.14.1
types-requests==2.32.4.20250809
overrides
boto3
botocore
grpcio
grpcio-status
grpcio-tools
tenacity==9.0.0
tree-sitter-java==0.21.0
tree-sitter-javascript==0.21.4
writer-sdk==2.1.0
anthropic==0.49.0
rsa
shapely
google-crc32c
docstring-parser
google-resumable-media
grpc-google-iam-v1
google-genai
google-cloud-core
google-cloud-resource-manager
google-cloud-bigquery
jmespath
s3transfer
absl-py
accelerate>=0.19.0
cpm_kernels
datasets<=3.6.0,>=2.12.0
einops>=0.5.0
evaluate>=0.3.0
func_timeout
fuzzywuzzy
gradio-client
h5py
httpx
immutabledict
importlib-metadata
jieba
json5
jsonlines
mmengine-lite
nltk>=3.7
numpy<2.0.0,>=1.23.4
openai
opencv-python-headless
orjson
pandas
plotly
prettytable
protobuf
pyext==0.5
python-Levenshtein
rank_bm25==0.2.2
rapidfuzz
requests>=2.31.0
retrying
rich
rouge
rouge_chinese
rouge_score
sacrebleu
scikit_learn>=1.5.0
scipy
seaborn
tabulate
tiktoken
timeout_decorator
tokenizers
torch
tqdm>=4.64.1
transformers
tree-sitter==0.21.3
tree_sitter_languages>=1.10.2
typer
Pillow==11.2.1
janus

[full]
human-eval
math-verify==0.5.2
langdetect
openai
qwen_vl_utils==0.0.14
torchvision
bfcl_eval==2025.6.14
annotated-types==0.7.0
anyio==4.10.0
argcomplete==3.6.2
black==25.1.0
cachetools==5.5.2
certifi==2025.8.3
charset-normalizer==3.4.3
click==8.2.1
cohere==5.13.3
datamodel-code-generator==0.25.7
distro==1.9.0
eval_type_backport==0.2.2
exceptiongroup==1.3.0
fastavro==1.12.0
filelock==3.19.1
fsspec==2024.12.0
genson==1.3.0
google-api-core==2.25.1
google-auth==2.40.3
google-cloud-aiplatform==1.93.1
google-cloud-storage==2.19.0
googleapis-common-protos==1.70.0
h11==0.16.0
hf-xet
httpcore==1.0.9
httpx-sse==0.4.0
huggingface-hub
idna==3.10
inflect==5.6.2
isort==5.13.2
jiter==0.10.0
Jinja2==3.1.6
Markdown-it-py==4.0.0
MarkupSafe==3.0.2
mdurl==0.1.2
mistralai==1.7.0
mypy_extensions==1.1.0
packaging==25.0
parameterized==0.9.0
pathspec==0.12.1
platformdirs==4.3.8
proto-plus==1.26.1
pydantic==2.11.7
pydantic_core==2.33.2
PyASN1==0.6.1
pyasn1_modules==0.4.2
Pygments==2.19.2
python-dateutil==2.9.0.post0
python-dotenv==1.1.1
pytz==2025.2
PyYAML==6.0.2
shellingham==1.5.4
six==1.17.0
sniffio==1.3.1
toml==0.10.2
tomli==2.2.1
typing-inspection==0.4.1
typing_extensions==4.14.1
types-requests==2.32.4.20250809
overrides
boto3
botocore
grpcio
grpcio-status
grpcio-tools
tenacity==9.0.0
tree-sitter-java==0.21.0
tree-sitter-javascript==0.21.4
writer-sdk==2.1.0
anthropic==0.49.0
rsa
shapely
google-crc32c
docstring-parser
google-resumable-media
grpc-google-iam-v1
google-genai
google-cloud-core
google-cloud-resource-manager
google-cloud-bigquery
jmespath
s3transfer
absl-py
accelerate>=0.19.0
cpm_kernels
datasets<=3.6.0,>=2.12.0
einops>=0.5.0
evaluate>=0.3.0
func_timeout
fuzzywuzzy
gradio-client
h5py
httpx
immutabledict
importlib-metadata
jieba
json5
jsonlines
mmengine-lite
nltk>=3.7
numpy<2.0.0,>=1.23.4
opencv-python-headless
orjson
pandas
plotly
prettytable
protobuf
pyext==0.5
python-Levenshtein
rank_bm25==0.2.2
rapidfuzz
requests>=2.31.0
retrying
rich
rouge
rouge_chinese
rouge_score
sacrebleu
scikit_learn>=1.5.0
scipy
seaborn
tabulate
tiktoken
timeout_decorator
tokenizers
torch
tqdm>=4.64.1
transformers
tree-sitter==0.21.3
tree_sitter_languages>=1.10.2
typer
Pillow==11.2.1
janus

[hf_vl]
qwen_vl_utils==0.0.14
torchvision
absl-py
accelerate>=0.19.0
cpm_kernels
datasets<=3.6.0,>=2.12.0
einops>=0.5.0
evaluate>=0.3.0
func_timeout
fuzzywuzzy
gradio-client
h5py
httpx
immutabledict
importlib-metadata
jieba
json5
jsonlines
mmengine-lite
nltk>=3.7
numpy<2.0.0,>=1.23.4
openai
opencv-python-headless
orjson
pandas
plotly
prettytable
protobuf
pyext==0.5
python-Levenshtein
rank_bm25==0.2.2
rapidfuzz
requests>=2.31.0
retrying
rich
rouge
rouge_chinese
rouge_score
sacrebleu
scikit_learn>=1.5.0
scipy
seaborn
tabulate
tiktoken
timeout_decorator
tokenizers
torch
tqdm>=4.64.1
transformers
tree-sitter==0.21.3
tree_sitter_languages>=1.10.2
typer
Pillow==11.2.1
janus

[ocrbench_v2]
apted
distance
editdistance
ipdb
jieba
Levenshtein
lxml
nltk
numpy
Polygon3
tqdm
zss
absl-py
accelerate>=0.19.0
cpm_kernels
datasets<=3.6.0,>=2.12.0
einops>=0.5.0
evaluate>=0.3.0
func_timeout
fuzzywuzzy
gradio-client
h5py
httpx
immutabledict
importlib-metadata
json5
jsonlines
mmengine-lite
nltk>=3.7
numpy<2.0.0,>=1.23.4
openai
opencv-python-headless
orjson
pandas
plotly
prettytable
protobuf
pyext==0.5
python-Levenshtein
rank_bm25==0.2.2
rapidfuzz
requests>=2.31.0
retrying
rich
rouge
rouge_chinese
rouge_score
sacrebleu
scikit_learn>=1.5.0
scipy
seaborn
tabulate
tiktoken
timeout_decorator
tokenizers
torch
tqdm>=4.64.1
transformers
tree-sitter==0.21.3
tree_sitter_languages>=1.10.2
typer
Pillow==11.2.1
janus
