Supported aten operators

Note

This table and page are auto generated from 'a script' that dig into PyTorch. Version targetted is: 'v2.7.1'. file was generated the 28 Jul 2025.

Warning

Take these informations with a grain of salt as this is referencing operators that may never appear in torch IR graph traced by torch_to_nnef (because remapped to others more generic). Also some uncommon operators are very rare in models, hence support may be lacking. SONOS only maintains operators 'per need basis', but contributions are always wecome see how.

'is core' column refers to this PyTorch IR documentation page

We filter-out from from observed operators 'backward' and 'sym' one's which are unwanted in inference engine. Also in place operations are merged with memory allocated activations as this is inference implementation detail.

TractNNEFONNX

Total matched operators in torch_to_nnef compared to:

• core PyTorch opset:

114/138

• and support from full aten:::

186/862

(total registered aten operators in t2n being 217)

aten name	aliases	can in-place	is core	translated
abs	absolute	✅	✅	✅
acos	arccos	✅	✅	✅
acosh	arccosh	✅	✅	✅
adaptive_avg_pool1d		❌	✅	✅
add		✅	✅	✅
addmm		✅	✅	✅
alias		❌	✅	✅
amax		❌	✅	✅
amin		❌	✅	✅
any		❌	✅	✅
arange		❌	✅	✅
argmax		❌	✅	✅
argmin		❌	✅	✅
as_strided		✅	✅	❌
asin	arcsin	✅	✅	✅
asinh	arcsinh	✅	✅	✅
atan	arctan	✅	✅	✅
atan2	arctan2	✅	✅	✅
atanh	arctanh	✅	✅	✅
avg_pool1d		❌	✅	✅
avg_pool2d		❌	✅	✅
avg_pool3d		❌	✅	✅
bitwise_and		✅	✅	✅
bitwise_not		✅	✅	✅
bitwise_or		✅	✅	✅
bitwise_xor		✅	✅	✅
bmm		❌	✅	✅
cat	concat, concatenate	❌	✅	✅
ceil		✅	✅	✅
clamp	clip	✅	✅	✅
clone		❌	✅	✅
col2im		❌	✅	❌
constant_pad_nd		❌	✅	✅
convolution		❌	✅	✅
copy		✅	✅	✅
cos		✅	✅	✅
cosh		✅	✅	✅
cumsum		✅	✅	❌
diagonal		❌	✅	❌
div	divide, true_divide	✅	✅	✅
elu		✅	✅	✅
embedding		❌	✅	✅
empty		❌	✅	❌
empty_strided		❌	✅	❌
eq		✅	✅	✅
erf	special_erf	✅	✅	✅
exp		✅	✅	✅
expand		❌	✅	✅
expm1	special_expm1	✅	✅	✅
fill		✅	✅	✅
flip		❌	✅	❌
floor		✅	✅	✅
fmod		✅	✅	✅
full		❌	✅	✅
full_like		❌	✅	✅
gather		❌	✅	✅
ge	greater_equal	✅	✅	✅
gelu		✅	✅	✅
grid_sampler_2d		❌	✅	❌
gt	greater	✅	✅	✅
hardtanh		✅	✅	✅
index		❌	✅	✅
index_put		✅	✅	❌
index_select		❌	✅	✅
isinf		❌	✅	❌
isnan		❌	✅	❌
le	less_equal	✅	✅	✅
leaky_relu		✅	✅	✅
log		✅	✅	✅
log10		✅	✅	✅
log1p	special_log1p	✅	✅	✅
log2		✅	✅	✅
logical_and		✅	✅	✅
logical_not		✅	✅	✅
logical_or		✅	✅	✅
logical_xor		✅	✅	✅
lt	less	✅	✅	✅
masked_scatter		✅	✅	❌
max		❌	✅	✅
max_pool2d_with_indices		❌	✅	❌
max_pool3d_with_indices		❌	✅	❌
maximum		❌	✅	✅
mean		❌	✅	✅
min		❌	✅	✅
minimum		❌	✅	✅
mm		❌	✅	✅
mul	multiply	✅	✅	✅
native_dropout		❌	✅	✅
native_group_norm		❌	✅	✅
native_layer_norm		❌	✅	✅
ne	not_equal	✅	✅	✅
neg	negative	✅	✅	✅
nonzero		❌	✅	❌
permute		❌	✅	✅
pow		✅	✅	✅
prod		❌	✅	✅
rand		❌	✅	❌
randn		❌	✅	❌
randperm		❌	✅	❌
reciprocal		✅	✅	✅
reflection_pad1d		❌	✅	✅
reflection_pad2d		❌	✅	✅
reflection_pad3d		❌	✅	✅
relu		✅	✅	✅
remainder		✅	✅	✅
repeat		❌	✅	✅
replication_pad2d		❌	✅	✅
replication_pad3d		❌	✅	✅
round	special_round	✅	✅	✅
rsqrt		✅	✅	✅
scalar_tensor		❌	✅	✅
scatter		✅	✅	✅
scatter_add		✅	✅	❌
scatter_reduce		✅	✅	❌
select		❌	✅	✅
select_scatter		❌	✅	❌
sigmoid	special_expit	✅	✅	✅
sign		✅	✅	✅
sin		✅	✅	✅
sinh		✅	✅	✅
slice		❌	✅	✅
slice_scatter		❌	✅	❌
sort		❌	✅	✅
split_with_sizes		❌	✅	✅
sqrt		✅	✅	✅
squeeze		✅	✅	✅
sub	subtract	✅	✅	✅
sum		❌	✅	✅
tan		✅	✅	✅
tanh		✅	✅	✅
topk		❌	✅	✅
trunc	fix	✅	✅	✅
unsqueeze		✅	✅	✅
upsample_bilinear2d		❌	✅	❌
upsample_nearest2d		❌	✅	❌
var		❌	✅	✅
view		❌	✅	✅
where		❌	✅	✅
adaptive_avg_pool2d		❌	-	✅
adaptive_avg_pool3d		❌	-	✅
adaptive_max_pool1d		❌	-	✅
adaptive_max_pool2d		❌	-	✅
adaptive_max_pool3d		❌	-	✅
addbmm		✅	-	❌
addcdiv		❌	-	❌
addcmul		❌	-	❌
addmv		✅	-	❌
addr		✅	-	❌
affine_grid_generator		❌	-	❌
alias_copy		❌	-	❌
align_as		❌	-	❌
align_tensors		❌	-	❌
align_to		❌	-	❌
all		❌	-	✅
all_gather_into_tensor		❌	-	❌
all_reduce		❌	-	❌
allclose		❌	-	❌
alpha_dropout		✅	-	❌
aminmax		❌	-	❌
angle		❌	-	❌
append		❌	-	❌
argsort		❌	-	✅
argwhere		❌	-	❌
as_strided_copy		❌	-	❌
as_strided_scatter		❌	-	❌
as_tensor		❌	-	❌
atleast_1d		❌	-	❌
atleast_2d		❌	-	❌
atleast_3d		❌	-	❌
baddbmm		✅	-	✅
bartlett_window		❌	-	❌
batch_norm		❌	-	✅
batch_norm_elemt		❌	-	❌
batch_norm_gather_stats		❌	-	❌
batch_norm_gather_stats_with_counts		❌	-	❌
batch_norm_stats		❌	-	❌
batch_norm_update_stats		❌	-	❌
bernoulli		✅	-	❌
bilinear		❌	-	❌
bin		❌	-	❌
binary_cross_entropy		❌	-	❌
binary_cross_entropy_with_logits		❌	-	❌
bincount		❌	-	❌
binomial		❌	-	❌
bitwise_left_shift		✅	-	❌
bitwise_right_shift		✅	-	❌
blackman_window		❌	-	❌
block_diag		❌	-	❌
broadcast_tensors		❌	-	❌
broadcast_to		❌	-	❌
bucketize		❌	-	❌
can_cast		❌	-	❌
capitalize		❌	-	❌
cartesian_prod		❌	-	❌
cauchy		✅	-	❌
ccol_indices		❌	-	❌
ccol_indices_copy		❌	-	❌
cdist		❌	-	❌
celu		❌	-	❌
center		❌	-	❌
chain_matmul		❌	-	❌
chalf		❌	-	❌
channel_shuffle		❌	-	❌
cholesky		❌	-	❌
cholesky_inverse		❌	-	❌
cholesky_solve		❌	-	❌
choose_qparams_optimized		❌	-	❌
chr		❌	-	❌
chunk		❌	-	✅
clamp_max		✅	-	✅
clamp_min		✅	-	✅
clear		❌	-	❌
coalesce		❌	-	❌
col_indices		❌	-	❌
col_indices_copy		❌	-	❌
column_stack		❌	-	❌
combinations		❌	-	❌
complex		❌	-	❌
confirmed_by_owner		❌	-	❌
conj		❌	-	❌
conj_physical		✅	-	❌
contiguous		❌	-	✅
conv		❌	-	❌
conv1d		❌	-	✅
conv2d		❌	-	✅
conv3d		❌	-	✅
conv_depthwise3d		❌	-	❌
conv_tbc		❌	-	❌
conv_transpose1d		❌	-	❌
conv_transpose2d		❌	-	❌
conv_transpose3d		❌	-	❌
convolution_overrideable		❌	-	❌
convrelu		❌	-	❌
copy_sparse_to_sparse		✅	-	❌
copy_to		❌	-	❌
copysign		✅	-	❌
corrcoef		❌	-	❌
cosine_embedding_loss		❌	-	❌
cosine_similarity		❌	-	❌
count		❌	-	❌
count_nonzero		❌	-	❌
cpu		❌	-	❌
cross		❌	-	❌
cross_entropy_loss		❌	-	❌
crow_indices		❌	-	❌
crow_indices_copy		❌	-	❌
ctc_loss		❌	-	❌
cuda		❌	-	❌
cudnn_affine_grid_generator		❌	-	❌
cudnn_batch_norm		❌	-	❌
cudnn_convolution		❌	-	❌
cudnn_convolution_add_relu		❌	-	❌
cudnn_convolution_relu		❌	-	❌
cudnn_convolution_transpose		❌	-	❌
cudnn_grid_sampler		❌	-	❌
cudnn_is_acceptable		❌	-	❌
cummax		❌	-	❌
cummin		❌	-	❌
cumprod		✅	-	❌
cumulative_trapezoid		❌	-	❌
data		❌	-	❌
deg2rad		❌	-	❌
degrees		❌	-	❌
dense_dim		❌	-	❌
dequantize		❌	-	✅
detach		✅	-	✅
detach_copy		❌	-	❌
device		❌	-	❌
diag		❌	-	❌
diag_embed		❌	-	❌
diagflat		❌	-	❌
diagonal_copy		❌	-	❌
diagonal_scatter		❌	-	❌
dict		❌	-	❌
diff		❌	-	❌
digamma	special_digamma, special_psi	✅	-	❌
dim		❌	-	❌
dist		❌	-	❌
divmod		❌	-	❌
dot		❌	-	❌
dropout		✅	-	✅
dsplit		❌	-	❌
dstack		❌	-	❌
dtype		❌	-	❌
eig		❌	-	❌
einsum		❌	-	✅
element_size		❌	-	❌
embedding_bag		❌	-	❌
embedding_renorm		✅	-	❌
empty_like		❌	-	✅
empty_permuted		❌	-	❌
empty_quantized		❌	-	❌
enable_grad		❌	-	❌
endswith		❌	-	❌
equal		❌	-	❌
erfc	special_erfc	✅	-	❌
erfinv	special_erfinv	✅	-	❌
exp2	special_exp2	✅	-	❌
expand_as		❌	-	❌
expand_copy		❌	-	❌
expandtabs		❌	-	❌
exponential		✅	-	❌
extend		❌	-	❌
eye		❌	-	❌
fabs		❌	-	❌
factorial		❌	-	❌
fake_quantize_per_channel_affine		❌	-	❌
fake_quantize_per_channel_affine_cachemask		❌	-	❌
fake_quantize_per_tensor_affine		❌	-	❌
fake_quantize_per_tensor_affine_cachemask		❌	-	❌
feature_alpha_dropout		✅	-	❌
feature_dropout		✅	-	❌
fft_fftfreq		❌	-	❌
fft_ihfft2		❌	-	❌
fft_ihfftn		❌	-	❌
fft_irfftn		❌	-	❌
fft_rfftfreq		❌	-	❌
fft_rfftn		❌	-	❌
fill_diagonal_		❌	-	❌
find		❌	-	❌
flatten		❌	-	✅
flatten_dense_tensors		❌	-	❌
fliplr		❌	-	❌
flipud		❌	-	❌
float_power		✅	-	❌
floor_divide		✅	-	✅
floordiv		❌	-	❌
fmax		❌	-	❌
fmin		❌	-	❌
foo		❌	-	❌
fork		❌	-	❌
format		❌	-	❌
frac		❌	-	❌
fractional_max_pool2d		❌	-	❌
fractional_max_pool3d		❌	-	❌
frexp		❌	-	❌
frobenius_norm		❌	-	❌
from_file		❌	-	❌
fused_moving_avg_obs_fake_quant		❌	-	❌
gamma		❌	-	❌
gcd		✅	-	❌
geometric		✅	-	❌
geqrf		❌	-	❌
get		❌	-	❌
get_autocast_dtype		❌	-	❌
get_device		❌	-	❌
get_gradients		❌	-	❌
get_pool_ceil_padding		❌	-	❌
getelem		❌	-	❌
glu		❌	-	✅
glu_jvp		❌	-	❌
grad		❌	-	❌
grid_sampler		❌	-	❌
grid_sampler_3d		❌	-	❌
group_norm		❌	-	✅
gru		❌	-	❌
gru_cell		❌	-	❌
hamming_window		❌	-	❌
hann_window		❌	-	❌
hardshrink		✅	-	❌
hardsigmoid		✅	-	❌
hardswish		✅	-	✅
has_torch_function		❌	-	❌
hash		❌	-	❌
heaviside		❌	-	❌
hex		❌	-	❌
hinge_embedding_loss		❌	-	❌
histc		❌	-	❌
histogram		❌	-	❌
histogramdd		❌	-	❌
hsplit		❌	-	❌
hspmm		❌	-	❌
hstack		❌	-	✅
huber_loss		❌	-	❌
hypot		✅	-	❌
i0	special_i0	✅	-	❌
igamma	special_gammainc	✅	-	❌
igammac	special_gammaincc	✅	-	❌
iinfo		❌	-	❌
im2col		❌	-	❌
imag		❌	-	❌
index_add		❌	-	❌
index_copy		❌	-	❌
index_fill		❌	-	❌
index_put_impl_		❌	-	❌
index_reduce		✅	-	❌
indices		❌	-	❌
indices_copy		❌	-	❌
initial_seed		❌	-	❌
inner		❌	-	❌
insert		❌	-	❌
instance_norm		❌	-	❌
int_repr		❌	-	❌
is_autocast_cpu_enabled		❌	-	❌
is_autocast_enabled		❌	-	❌
is_coalesced		❌	-	❌
is_complex		❌	-	❌
is_conj		❌	-	❌
is_contiguous		❌	-	❌
is_cuda		❌	-	❌
is_floating_point		❌	-	❌
is_grad_enabled		❌	-	❌
is_leaf		❌	-	❌
is_non_overlapping_and_dense		❌	-	❌
is_nonzero		❌	-	❌
is_owner		❌	-	❌
is_pinned		❌	-	❌
is_same_size		❌	-	❌
is_scripting		❌	-	❌
is_set_to		❌	-	❌
is_signed		❌	-	❌
is_strides_like_format		❌	-	❌
isalnum		❌	-	❌
isalpha		❌	-	❌
isclose		❌	-	❌
isdecimal		❌	-	❌
isdigit		❌	-	❌
isfinite		❌	-	❌
isidentifier		❌	-	❌
isin		❌	-	❌
islower		❌	-	❌
isneginf		❌	-	❌
isnumeric		❌	-	❌
isposinf		❌	-	❌
isprintable		❌	-	❌
isreal		❌	-	❌
isspace		❌	-	❌
istft		❌	-	❌
istitle		❌	-	❌
isupper		❌	-	❌
item		❌	-	❌
items		❌	-	❌
join		❌	-	❌
kaiser_window		❌	-	❌
keys		❌	-	❌
kl_div		❌	-	❌
kron		❌	-	❌
kthvalue		❌	-	❌
l1_loss		❌	-	❌
layer_norm		❌	-	✅
lcm		✅	-	❌
ldexp		✅	-	❌
len		❌	-	❌
lerp		✅	-	❌
lgamma		✅	-	❌
lift		❌	-	❌
lift_fresh		❌	-	❌
lift_fresh_copy		❌	-	❌
linalg_cholesky_ex		❌	-	❌
linalg_cond		❌	-	❌
linalg_cross		❌	-	❌
linalg_det	det	❌	-	❌
linalg_diagonal		❌	-	❌
linalg_eig		❌	-	❌
linalg_eigh		❌	-	❌
linalg_eigvals		❌	-	❌
linalg_eigvalsh		❌	-	❌
linalg_householder_product	orgqr	❌	-	❌
linalg_inv	inverse	❌	-	❌
linalg_inv_ex		❌	-	❌
linalg_ldl_factor_ex		❌	-	❌
linalg_ldl_solve		❌	-	❌
linalg_lstsq		❌	-	❌
linalg_lu		❌	-	❌
linalg_lu_factor_ex		❌	-	❌
linalg_lu_solve		❌	-	❌
linalg_matrix_exp	matrix_exp	❌	-	❌
linalg_matrix_norm		❌	-	❌
linalg_matrix_power	matrix_power	❌	-	❌
linalg_matrix_rank		❌	-	❌
linalg_norm		❌	-	✅
linalg_pinv		❌	-	❌
linalg_qr		❌	-	❌
linalg_slogdet		❌	-	❌
linalg_solve		❌	-	❌
linalg_solve_triangular		❌	-	❌
linalg_svd		❌	-	❌
linalg_tensorinv		❌	-	❌
linalg_tensorsolve		❌	-	❌
linalg_vector_norm		❌	-	✅
linear		❌	-	✅
linspace		❌	-	❌
list		❌	-	❌
list_with_default		❌	-	❌
ljust		❌	-	❌
local_value		❌	-	❌
log_normal		✅	-	❌
log_sigmoid		❌	-	❌
log_sigmoid_forward		❌	-	❌
log_softmax	special_log_softmax	❌	-	✅
logaddexp		❌	-	❌
logaddexp2		❌	-	❌
logcumsumexp		❌	-	❌
logdet		❌	-	❌
logit	special_logit	❌	-	❌
logspace		❌	-	❌
logsumexp	special_logsumexp	❌	-	❌
lower		❌	-	❌
lstm		❌	-	❌
lstm_cell		❌	-	❌
lstrip		❌	-	❌
lstsq		❌	-	❌
lu_solve		❌	-	❌
lu_unpack		❌	-	❌
mH	adjoint	❌	-	❌
mT		❌	-	❌
manual_seed		❌	-	❌
margin_ranking_loss		❌	-	❌
masked_fill		✅	-	✅
masked_select		❌	-	❌
mathremainder		❌	-	❌
matmul	linalg_matmul	❌	-	✅
matrix_H		❌	-	❌
matrix_rank		❌	-	❌
max_pool1d		❌	-	✅
max_pool1d_with_indices		❌	-	❌
max_pool2d		❌	-	✅
max_pool3d		❌	-	✅
max_unpool2d		❌	-	❌
max_unpool3d		❌	-	❌
median		❌	-	❌
meshgrid		❌	-	❌
miopen_batch_norm		❌	-	❌
miopen_convolution		❌	-	❌
miopen_convolution_add_relu		❌	-	❌
miopen_convolution_relu		❌	-	❌
miopen_convolution_transpose		❌	-	❌
miopen_depthwise_convolution		❌	-	❌
miopen_rnn		❌	-	❌
mish		❌	-	❌
mkldnn_adaptive_avg_pool2d		❌	-	❌
mkldnn_convolution		❌	-	❌
mkldnn_linear		❌	-	❌
mkldnn_max_pool2d		❌	-	❌
mkldnn_max_pool3d		❌	-	❌
mkldnn_reorder_conv2d_weight		❌	-	❌
mkldnn_reorder_conv3d_weight		❌	-	❌
mkldnn_rnn_layer		❌	-	❌
mode		❌	-	❌
modf		❌	-	❌
movedim	moveaxis	❌	-	❌
mps_linear		❌	-	❌
mse_loss		❌	-	❌
msort		❌	-	❌
multi_margin_loss		❌	-	❌
multilabel_margin_loss		❌	-	❌
multilabel_margin_loss_forward		❌	-	❌
multinomial		❌	-	❌
mv		❌	-	❌
mvlgamma	special_multigammaln	❌	-	❌
nan_to_num		✅	-	❌
nanmean		❌	-	❌
nanmedian		❌	-	❌
nanquantile		❌	-	❌
nansum		❌	-	❌
narrow		❌	-	✅
narrow_copy		❌	-	❌
native_batch_norm		❌	-	❌
native_channel_shuffle		❌	-	❌
native_multi_head_self_attention		❌	-	❌
native_norm		❌	-	❌
neq		❌	-	❌
nested_to_padded_tensor		❌	-	❌
new_empty		❌	-	❌
new_empty_strided		❌	-	❌
new_full		❌	-	❌
new_ones		❌	-	❌
new_zeros		❌	-	✅
nextafter		✅	-	❌
nll_loss		❌	-	❌
nll_loss2d		❌	-	❌
nll_loss_forward		❌	-	❌
nll_loss_nd		❌	-	❌
node		❌	-	❌
nonzero_numpy		❌	-	❌
nonzero_static		❌	-	❌
norm		❌	-	✅
norm_except_dim		❌	-	❌
normal		✅	-	❌
normal_functional		❌	-	❌
nuclear_norm		❌	-	❌
numel		❌	-	✅
numpy_T		❌	-	❌
oct		❌	-	❌
one_hot		❌	-	❌
ones		❌	-	✅
ones_like		❌	-	✅
op		❌	-	❌
op_name		❌	-	❌
ord		❌	-	❌
ormqr		❌	-	❌
outer	ger	❌	-	❌
output_nr		❌	-	❌
owner		❌	-	❌
owner_name		❌	-	❌
pad		❌	-	✅
pad_sequence		❌	-	❌
pairwise_distance		❌	-	❌
partition		❌	-	❌
pdist		❌	-	❌
percentFormat		❌	-	❌
permute_copy		❌	-	❌
pin_memory		❌	-	❌
pinv		❌	-	❌
pinverse		❌	-	❌
pixel_shuffle		❌	-	❌
pixel_unshuffle		❌	-	❌
pointwise_placeholder		❌	-	❌
poisson		❌	-	❌
poisson_nll_loss		❌	-	❌
polar		❌	-	❌
polygamma	special_polygamma	✅	-	❌
pop		❌	-	❌
popitem		❌	-	❌
positive		❌	-	❌
prelu		❌	-	✅
promote_types		❌	-	❌
put		✅	-	❌
q_per_channel_axis		❌	-	❌
q_per_channel_scales		❌	-	❌
q_per_channel_zero_points		❌	-	❌
q_scale		❌	-	❌
q_zero_point		❌	-	❌
qr		❌	-	❌
qscheme		❌	-	❌
quantile		❌	-	❌
quantize		❌	-	❌
quantize_per_channel		❌	-	❌
quantize_per_tensor		❌	-	✅
quantize_per_tensor_dynamic		❌	-	❌
quantized_batch_norm		❌	-	❌
quantized_gru		❌	-	❌
quantized_lstm		❌	-	❌
quantized_max_pool1d		❌	-	❌
quantized_max_pool2d		❌	-	❌
quantized_max_pool3d		❌	-	❌
rad2deg		❌	-	❌
radians		❌	-	❌
rand_like		❌	-	❌
randint		❌	-	❌
randint_like		❌	-	❌
randn_like		❌	-	❌
random		✅	-	❌
range		❌	-	❌
ravel		❌	-	❌
real		❌	-	❌
record_stream		❌	-	❌
reduce_scatter_tensor		❌	-	❌
refine_names		❌	-	❌
relu6		✅	-	✅
remove		❌	-	❌
rename		✅	-	❌
renorm		❌	-	❌
repeat_interleave		❌	-	✅
replace		❌	-	❌
replication_pad1d		❌	-	✅
requires_grad_		❌	-	❌
reshape		❌	-	✅
reshape_as		❌	-	❌
resize		✅	-	❌
resize_as_		❌	-	❌
resize_as_sparse		✅	-	❌
resolve_conj		❌	-	❌
resolve_neg		❌	-	❌
result_type		❌	-	❌
retain_grad		❌	-	❌
retains_grad		❌	-	❌
reverse		❌	-	❌
rfind		❌	-	❌
rindex		❌	-	❌
rjust		❌	-	❌
rnn_relu		❌	-	❌
rnn_relu_cell		❌	-	❌
rnn_tanh		❌	-	❌
rnn_tanh_cell		❌	-	❌
roll		❌	-	✅
rot90		❌	-	❌
row_indices		❌	-	❌
row_indices_copy		❌	-	❌
rowwise_prune		❌	-	❌
rpartition		❌	-	❌
rrelu		❌	-	❌
rrelu_with_noise		✅	-	❌
rrelu_with_noise_functional		❌	-	❌
rsplit		❌	-	❌
rstrip		❌	-	❌
rsub		❌	-	✅
save		❌	-	❌
scaled_dot_product_attention		❌	-	✅
searchsorted		❌	-	❌
seed		❌	-	❌
segment_reduce		❌	-	❌
select_copy		❌	-	❌
selu		✅	-	✅
set		✅	-	❌
set_data		❌	-	❌
set_grad_enabled		❌	-	❌
set_source_Tensor_storage_offset		❌	-	❌
setdefault		❌	-	❌
sgn		❌	-	❌
signbit		❌	-	❌
silu		✅	-	✅
sinc	special_sinc	❌	-	❌
size		❌	-	✅
slice_copy		❌	-	❌
slice_inverse		❌	-	❌
slow_conv3d		❌	-	❌
slow_conv3d_forward		❌	-	❌
slow_conv_dilated2d		❌	-	❌
slow_conv_dilated3d		❌	-	❌
slow_conv_transpose2d		❌	-	❌
slow_conv_transpose3d		❌	-	❌
smm		❌	-	❌
smooth_l1_loss		❌	-	❌
soft_margin_loss		❌	-	❌
softmax	special_softmax	❌	-	✅
softplus		❌	-	✅
softshrink		❌	-	❌
solve		❌	-	❌
sorted		❌	-	❌
sparse_compressed_tensor		❌	-	❌
sparse_coo_tensor		❌	-	❌
sparse_dim		❌	-	❌
sparse_mask		❌	-	❌
sparse_resize		✅	-	❌
sparse_resize_and_clear		✅	-	❌
sparse_sampled_addmm		❌	-	❌
special_airy_ai		❌	-	❌
special_bessel_j0		❌	-	❌
special_bessel_j1		❌	-	❌
special_bessel_y0		❌	-	❌
special_bessel_y1		❌	-	❌
special_chebyshev_polynomial_t		❌	-	❌
special_chebyshev_polynomial_u		❌	-	❌
special_chebyshev_polynomial_v		❌	-	❌
special_chebyshev_polynomial_w		❌	-	❌
special_entr		❌	-	❌
special_erfcx		❌	-	❌
special_hermite_polynomial_h		❌	-	❌
special_hermite_polynomial_he		❌	-	❌
special_i0e		❌	-	❌
special_i1		❌	-	❌
special_i1e		❌	-	❌
special_laguerre_polynomial_l		❌	-	❌
special_legendre_polynomial_p		❌	-	❌
special_log_ndtr		❌	-	❌
special_modified_bessel_i0		❌	-	❌
special_modified_bessel_i1		❌	-	❌
special_modified_bessel_k0		❌	-	❌
special_modified_bessel_k1		❌	-	❌
special_ndtr		❌	-	❌
special_ndtri		❌	-	❌
special_scaled_modified_bessel_k0		❌	-	❌
special_scaled_modified_bessel_k1		❌	-	❌
special_shifted_chebyshev_polynomial_t		❌	-	❌
special_shifted_chebyshev_polynomial_u		❌	-	❌
special_shifted_chebyshev_polynomial_v		❌	-	❌
special_shifted_chebyshev_polynomial_w		❌	-	❌
special_spherical_bessel_j0		❌	-	❌
special_xlog1py		❌	-	❌
special_zeta		❌	-	❌
split		❌	-	❌
split_copy		❌	-	❌
split_with_sizes_copy		❌	-	❌
splitlines		❌	-	❌
square		✅	-	❌
sspaddmm		❌	-	❌
stack		❌	-	✅
startswith		❌	-	❌
std		❌	-	❌
std_mean		❌	-	❌
stft		❌	-	✅
storage_offset		❌	-	❌
str		❌	-	❌
stride		❌	-	❌
strip		❌	-	❌
sum_to		❌	-	❌
svd		❌	-	❌
swapcase		❌	-	❌
symbolic_b		❌	-	❌
symeig		❌	-	❌
t		✅	-	❌
take		❌	-	❌
take_along_dim		❌	-	❌
tanhshrink		❌	-	❌
tensor		❌	-	❌
tensor_split		❌	-	❌
tensordot		❌	-	❌
test		❌	-	❌
test_symbol		❌	-	❌
test_vartype		❌	-	❌
test_vartype2		❌	-	❌
thnn_conv2d		❌	-	❌
thnn_conv2d_forward		❌	-	❌
threshold		❌	-	❌
tile		❌	-	❌
title		❌	-	❌
to		❌	-	✅
to_dense		❌	-	❌
to_here		❌	-	❌
to_mkldnn		❌	-	❌
to_padded_tensor		❌	-	❌
trace		❌	-	❌
transpose	swapaxes, swapdims	✅	-	✅
transpose_copy		❌	-	❌
trapezoid		❌	-	❌
trapz		❌	-	❌
triangular_solve		❌	-	❌
tril		❌	-	✅
tril_indices		❌	-	❌
triplet_margin_loss		❌	-	❌
triu		❌	-	✅
triu_indices		❌	-	❌
type_as		❌	-	✅
unbind		❌	-	✅
unbind_copy		❌	-	❌
unflatten		❌	-	✅
unflatten_dense_tensors		❌	-	❌
unfold		❌	-	❌
unfold_copy		❌	-	❌
uniform		✅	-	❌
unique_consecutive		❌	-	❌
unique_dim		❌	-	❌
unique_dim_consecutive		❌	-	❌
unknown		❌	-	❌
unsafe_chunk		❌	-	❌
unsafe_split		❌	-	❌
unsafe_split_with_sizes		❌	-	❌
unsqueeze_copy		❌	-	❌
update		❌	-	❌
upper		❌	-	❌
upsample		❌	-	❌
upsample_bicubic2d		❌	-	❌
upsample_linear1d		❌	-	❌
upsample_nearest1d		❌	-	❌
upsample_nearest3d		❌	-	❌
upsample_trilinear3d		❌	-	❌
values		❌	-	❌
values_copy		❌	-	❌
vander		❌	-	❌
var_mean		❌	-	❌
vdot		❌	-	❌
view_as		❌	-	❌
view_as_complex		❌	-	✅
view_as_complex_copy		❌	-	❌
view_as_real		❌	-	✅
view_as_real_copy		❌	-	❌
view_copy		❌	-	❌
view_expand_placeholder		❌	-	❌
vsplit		❌	-	❌
vstack	row_stack	❌	-	✅
wait		❌	-	❌
wait_tensor		❌	-	❌
warn		❌	-	❌
warns		❌	-	❌
wrapped_linear_prepack		❌	-	❌
wrapped_quantized_linear_prepacked		❌	-	❌
xlogy	special_xlogy	❌	-	❌
zero		✅	-	❌
zeros		❌	-	✅
zeros_like		❌	-	✅
zfill		❌	-	❌

Total matched operators in builtin PyTorch ONNX support based on this page compared to:

• core PyTorch opset:

123/138

• and support from full aten:::

320/862

(total registered aten operators in t2n being 348)

aten name	aliases	can in-place	is core	translated
abs	absolute	✅	✅	✅
acos	arccos	✅	✅	✅
acosh	arccosh	✅	✅	❌
adaptive_avg_pool1d		❌	✅	✅
add		✅	✅	✅
addmm		✅	✅	✅
alias		❌	✅	✅
amax		❌	✅	✅
amin		❌	✅	✅
any		❌	✅	✅
arange		❌	✅	✅
argmax		❌	✅	✅
argmin		❌	✅	✅
as_strided		✅	✅	✅
asin	arcsin	✅	✅	✅
asinh	arcsinh	✅	✅	❌
atan	arctan	✅	✅	✅
atan2	arctan2	✅	✅	✅
atanh	arctanh	✅	✅	❌
avg_pool1d		❌	✅	✅
avg_pool2d		❌	✅	✅
avg_pool3d		❌	✅	✅
bitwise_and		✅	✅	✅
bitwise_not		✅	✅	✅
bitwise_or		✅	✅	✅
bitwise_xor		✅	✅	❌
bmm		❌	✅	✅
cat	concat, concatenate	❌	✅	✅
ceil		✅	✅	✅
clamp	clip	✅	✅	✅
clone		❌	✅	✅
col2im		❌	✅	✅
constant_pad_nd		❌	✅	✅
convolution		❌	✅	✅
copy		✅	✅	❌
cos		✅	✅	✅
cosh		✅	✅	❌
cumsum		✅	✅	✅
diagonal		❌	✅	✅
div	divide, true_divide	✅	✅	✅
elu		✅	✅	✅
embedding		❌	✅	✅
empty		❌	✅	✅
empty_strided		❌	✅	❌
eq		✅	✅	✅
erf	special_erf	✅	✅	✅
exp		✅	✅	✅
expand		❌	✅	✅
expm1	special_expm1	✅	✅	❌
fill		✅	✅	✅
flip		❌	✅	✅
floor		✅	✅	✅
fmod		✅	✅	✅
full		❌	✅	✅
full_like		❌	✅	✅
gather		❌	✅	✅
ge	greater_equal	✅	✅	✅
gelu		✅	✅	✅
grid_sampler_2d		❌	✅	❌
gt	greater	✅	✅	✅
hardtanh		✅	✅	✅
index		❌	✅	✅
index_put		✅	✅	✅
index_select		❌	✅	✅
isinf		❌	✅	✅
isnan		❌	✅	✅
le	less_equal	✅	✅	✅
leaky_relu		✅	✅	✅
log		✅	✅	✅
log10		✅	✅	✅
log1p	special_log1p	✅	✅	✅
log2		✅	✅	✅
logical_and		✅	✅	✅
logical_not		✅	✅	✅
logical_or		✅	✅	✅
logical_xor		✅	✅	✅
lt	less	✅	✅	✅
masked_scatter		✅	✅	✅
max		❌	✅	✅
max_pool2d_with_indices		❌	✅	✅
max_pool3d_with_indices		❌	✅	✅
maximum		❌	✅	✅
mean		❌	✅	✅
min		❌	✅	✅
minimum		❌	✅	✅
mm		❌	✅	✅
mul	multiply	✅	✅	✅
native_dropout		❌	✅	✅
native_group_norm		❌	✅	❌
native_layer_norm		❌	✅	✅
ne	not_equal	✅	✅	✅
neg	negative	✅	✅	✅
nonzero		❌	✅	✅
permute		❌	✅	✅
pow		✅	✅	✅
prod		❌	✅	✅
rand		❌	✅	✅
randn		❌	✅	✅
randperm		❌	✅	❌
reciprocal		✅	✅	✅
reflection_pad1d		❌	✅	✅
reflection_pad2d		❌	✅	✅
reflection_pad3d		❌	✅	✅
relu		✅	✅	✅
remainder		✅	✅	✅
repeat		❌	✅	✅
replication_pad2d		❌	✅	✅
replication_pad3d		❌	✅	✅
round	special_round	✅	✅	✅
rsqrt		✅	✅	✅
scalar_tensor		❌	✅	✅
scatter		✅	✅	✅
scatter_add		✅	✅	✅
scatter_reduce		✅	✅	✅
select		❌	✅	✅
select_scatter		❌	✅	❌
sigmoid	special_expit	✅	✅	✅
sign		✅	✅	✅
sin		✅	✅	✅
sinh		✅	✅	❌
slice		❌	✅	✅
slice_scatter		❌	✅	❌
sort		❌	✅	✅
split_with_sizes		❌	✅	✅
sqrt		✅	✅	✅
squeeze		✅	✅	✅
sub	subtract	✅	✅	✅
sum		❌	✅	✅
tan		✅	✅	✅
tanh		✅	✅	✅
topk		❌	✅	✅
trunc	fix	✅	✅	❌
unsqueeze		✅	✅	✅
upsample_bilinear2d		❌	✅	✅
upsample_nearest2d		❌	✅	✅
var		❌	✅	✅
view		❌	✅	✅
where		❌	✅	✅
adaptive_avg_pool2d		❌	-	✅
adaptive_avg_pool3d		❌	-	✅
adaptive_max_pool1d		❌	-	✅
adaptive_max_pool2d		❌	-	✅
adaptive_max_pool3d		❌	-	✅
addbmm		✅	-	❌
addcdiv		❌	-	❌
addcmul		❌	-	✅
addmv		✅	-	❌
addr		✅	-	❌
affine_grid_generator		❌	-	✅
alias_copy		❌	-	❌
align_as		❌	-	❌
align_tensors		❌	-	❌
align_to		❌	-	❌
all		❌	-	✅
all_gather_into_tensor		❌	-	❌
all_reduce		❌	-	❌
allclose		❌	-	❌
alpha_dropout		✅	-	✅
aminmax		❌	-	✅
angle		❌	-	❌
append		❌	-	✅
argsort		❌	-	✅
argwhere		❌	-	❌
as_strided_copy		❌	-	❌
as_strided_scatter		❌	-	❌
as_tensor		❌	-	✅
atleast_1d		❌	-	✅
atleast_2d		❌	-	✅
atleast_3d		❌	-	✅
baddbmm		✅	-	✅
bartlett_window		❌	-	❌
batch_norm		❌	-	✅
batch_norm_elemt		❌	-	❌
batch_norm_gather_stats		❌	-	❌
batch_norm_gather_stats_with_counts		❌	-	❌
batch_norm_stats		❌	-	❌
batch_norm_update_stats		❌	-	❌
bernoulli		✅	-	✅
bilinear		❌	-	❌
bin		❌	-	❌
binary_cross_entropy		❌	-	❌
binary_cross_entropy_with_logits		❌	-	✅
bincount		❌	-	❌
binomial		❌	-	❌
bitwise_left_shift		✅	-	✅
bitwise_right_shift		✅	-	✅
blackman_window		❌	-	❌
block_diag		❌	-	❌
broadcast_tensors		❌	-	✅
broadcast_to		❌	-	✅
bucketize		❌	-	✅
can_cast		❌	-	❌
capitalize		❌	-	❌
cartesian_prod		❌	-	❌
cauchy		✅	-	❌
ccol_indices		❌	-	❌
ccol_indices_copy		❌	-	❌
cdist		❌	-	✅
celu		❌	-	✅
center		❌	-	❌
chain_matmul		❌	-	❌
chalf		❌	-	❌
channel_shuffle		❌	-	❌
cholesky		❌	-	❌
cholesky_inverse		❌	-	❌
cholesky_solve		❌	-	❌
choose_qparams_optimized		❌	-	❌
chr		❌	-	❌
chunk		❌	-	✅
clamp_max		✅	-	✅
clamp_min		✅	-	✅
clear		❌	-	❌
coalesce		❌	-	❌
col_indices		❌	-	❌
col_indices_copy		❌	-	❌
column_stack		❌	-	❌
combinations		❌	-	❌
complex		❌	-	❌
confirmed_by_owner		❌	-	❌
conj		❌	-	❌
conj_physical		✅	-	✅
contiguous		❌	-	✅
conv		❌	-	❌
conv1d		❌	-	✅
conv2d		❌	-	✅
conv3d		❌	-	✅
conv_depthwise3d		❌	-	❌
conv_tbc		❌	-	✅
conv_transpose1d		❌	-	✅
conv_transpose2d		❌	-	✅
conv_transpose3d		❌	-	✅
convolution_overrideable		❌	-	❌
convrelu		❌	-	❌
copy_sparse_to_sparse		✅	-	❌
copy_to		❌	-	❌
copysign		✅	-	❌
corrcoef		❌	-	❌
cosine_embedding_loss		❌	-	❌
cosine_similarity		❌	-	✅
count		❌	-	❌
count_nonzero		❌	-	❌
cpu		❌	-	❌
cross		❌	-	✅
cross_entropy_loss		❌	-	✅
crow_indices		❌	-	❌
crow_indices_copy		❌	-	❌
ctc_loss		❌	-	❌
cuda		❌	-	❌
cudnn_affine_grid_generator		❌	-	❌
cudnn_batch_norm		❌	-	❌
cudnn_convolution		❌	-	❌
cudnn_convolution_add_relu		❌	-	❌
cudnn_convolution_relu		❌	-	❌
cudnn_convolution_transpose		❌	-	❌
cudnn_grid_sampler		❌	-	❌
cudnn_is_acceptable		❌	-	❌
cummax		❌	-	❌
cummin		❌	-	❌
cumprod		✅	-	❌
cumulative_trapezoid		❌	-	❌
data		❌	-	❌
deg2rad		❌	-	❌
degrees		❌	-	❌
dense_dim		❌	-	❌
dequantize		❌	-	✅
detach		✅	-	✅
detach_copy		❌	-	❌
device		❌	-	❌
diag		❌	-	❌
diag_embed		❌	-	❌
diagflat		❌	-	❌
diagonal_copy		❌	-	❌
diagonal_scatter		❌	-	❌
dict		❌	-	❌
diff		❌	-	❌
digamma	special_digamma, special_psi	✅	-	❌
dim		❌	-	✅
dist		❌	-	❌
divmod		❌	-	❌
dot		❌	-	✅
dropout		✅	-	✅
dsplit		❌	-	❌
dstack		❌	-	❌
dtype		❌	-	❌
eig		❌	-	❌
einsum		❌	-	✅
element_size		❌	-	❌
embedding_bag		❌	-	✅
embedding_renorm		✅	-	✅
empty_like		❌	-	✅
empty_permuted		❌	-	❌
empty_quantized		❌	-	❌
enable_grad		❌	-	❌
endswith		❌	-	❌
equal		❌	-	❌
erfc	special_erfc	✅	-	❌
erfinv	special_erfinv	✅	-	❌
exp2	special_exp2	✅	-	❌
expand_as		❌	-	✅
expand_copy		❌	-	❌
expandtabs		❌	-	❌
exponential		✅	-	❌
extend		❌	-	❌
eye		❌	-	✅
fabs		❌	-	❌
factorial		❌	-	❌
fake_quantize_per_channel_affine		❌	-	✅
fake_quantize_per_channel_affine_cachemask		❌	-	❌
fake_quantize_per_tensor_affine		❌	-	✅
fake_quantize_per_tensor_affine_cachemask		❌	-	❌
feature_alpha_dropout		✅	-	✅
feature_dropout		✅	-	✅
fft_fftfreq		❌	-	❌
fft_ihfft2		❌	-	❌
fft_ihfftn		❌	-	❌
fft_irfftn		❌	-	❌
fft_rfftfreq		❌	-	❌
fft_rfftn		❌	-	❌
fill_diagonal_		❌	-	❌
find		❌	-	❌
flatten		❌	-	✅
flatten_dense_tensors		❌	-	❌
fliplr		❌	-	❌
flipud		❌	-	❌
float_power		✅	-	❌
floor_divide		✅	-	✅
floordiv		❌	-	✅
fmax		❌	-	❌
fmin		❌	-	❌
foo		❌	-	❌
fork		❌	-	❌
format		❌	-	❌
frac		❌	-	❌
fractional_max_pool2d		❌	-	❌
fractional_max_pool3d		❌	-	❌
frexp		❌	-	❌
frobenius_norm		❌	-	✅
from_file		❌	-	❌
fused_moving_avg_obs_fake_quant		❌	-	❌
gamma		❌	-	❌
gcd		✅	-	❌
geometric		✅	-	❌
geqrf		❌	-	❌
get		❌	-	❌
get_autocast_dtype		❌	-	❌
get_device		❌	-	❌
get_gradients		❌	-	❌
get_pool_ceil_padding		❌	-	❌
getelem		❌	-	❌
glu		❌	-	✅
glu_jvp		❌	-	❌
grad		❌	-	❌
grid_sampler		❌	-	✅
grid_sampler_3d		❌	-	❌
group_norm		❌	-	✅
gru		❌	-	✅
gru_cell		❌	-	❌
hamming_window		❌	-	❌
hann_window		❌	-	✅
hardshrink		✅	-	✅
hardsigmoid		✅	-	✅
hardswish		✅	-	✅
has_torch_function		❌	-	❌
hash		❌	-	❌
heaviside		❌	-	❌
hex		❌	-	❌
hinge_embedding_loss		❌	-	❌
histc		❌	-	❌
histogram		❌	-	❌
histogramdd		❌	-	❌
hsplit		❌	-	❌
hspmm		❌	-	❌
hstack		❌	-	✅
huber_loss		❌	-	❌
hypot		✅	-	❌
i0	special_i0	✅	-	❌
igamma	special_gammainc	✅	-	❌
igammac	special_gammaincc	✅	-	❌
iinfo		❌	-	❌
im2col		❌	-	✅
imag		❌	-	❌
index_add		❌	-	✅
index_copy		❌	-	✅
index_fill		❌	-	✅
index_put_impl_		❌	-	❌
index_reduce		✅	-	❌
indices		❌	-	❌
indices_copy		❌	-	❌
initial_seed		❌	-	❌
inner		❌	-	❌
insert		❌	-	✅
instance_norm		❌	-	✅
int_repr		❌	-	❌
is_autocast_cpu_enabled		❌	-	❌
is_autocast_enabled		❌	-	❌
is_coalesced		❌	-	❌
is_complex		❌	-	❌
is_conj		❌	-	❌
is_contiguous		❌	-	❌
is_cuda		❌	-	❌
is_floating_point		❌	-	✅
is_grad_enabled		❌	-	❌
is_leaf		❌	-	❌
is_non_overlapping_and_dense		❌	-	❌
is_nonzero		❌	-	❌
is_owner		❌	-	❌
is_pinned		❌	-	✅
is_same_size		❌	-	❌
is_scripting		❌	-	❌
is_set_to		❌	-	❌
is_signed		❌	-	❌
is_strides_like_format		❌	-	❌
isalnum		❌	-	❌
isalpha		❌	-	❌
isclose		❌	-	❌
isdecimal		❌	-	❌
isdigit		❌	-	❌
isfinite		❌	-	✅
isidentifier		❌	-	❌
isin		❌	-	❌
islower		❌	-	❌
isneginf		❌	-	❌
isnumeric		❌	-	❌
isposinf		❌	-	❌
isprintable		❌	-	❌
isreal		❌	-	❌
isspace		❌	-	❌
istft		❌	-	❌
istitle		❌	-	❌
isupper		❌	-	❌
item		❌	-	✅
items		❌	-	❌
join		❌	-	❌
kaiser_window		❌	-	❌
keys		❌	-	❌
kl_div		❌	-	✅
kron		❌	-	❌
kthvalue		❌	-	❌
l1_loss		❌	-	❌
layer_norm		❌	-	✅
lcm		✅	-	❌
ldexp		✅	-	❌
len		❌	-	✅
lerp		✅	-	✅
lgamma		✅	-	❌
lift		❌	-	✅
lift_fresh		❌	-	❌
lift_fresh_copy		❌	-	❌
linalg_cholesky_ex		❌	-	❌
linalg_cond		❌	-	❌
linalg_cross		❌	-	✅
linalg_det	det	❌	-	✅
linalg_diagonal		❌	-	❌
linalg_eig		❌	-	❌
linalg_eigh		❌	-	❌
linalg_eigvals		❌	-	❌
linalg_eigvalsh		❌	-	❌
linalg_householder_product	orgqr	❌	-	❌
linalg_inv	inverse	❌	-	❌
linalg_inv_ex		❌	-	❌
linalg_ldl_factor_ex		❌	-	❌
linalg_ldl_solve		❌	-	❌
linalg_lstsq		❌	-	❌
linalg_lu		❌	-	❌
linalg_lu_factor_ex		❌	-	❌
linalg_lu_solve		❌	-	❌
linalg_matrix_exp	matrix_exp	❌	-	❌
linalg_matrix_norm		❌	-	✅
linalg_matrix_power	matrix_power	❌	-	❌
linalg_matrix_rank		❌	-	❌
linalg_norm		❌	-	✅
linalg_pinv		❌	-	❌
linalg_qr		❌	-	❌
linalg_slogdet		❌	-	❌
linalg_solve		❌	-	❌
linalg_solve_triangular		❌	-	❌
linalg_svd		❌	-	❌
linalg_tensorinv		❌	-	❌
linalg_tensorsolve		❌	-	❌
linalg_vector_norm		❌	-	✅
linear		❌	-	✅
linspace		❌	-	✅
list		❌	-	✅
list_with_default		❌	-	❌
ljust		❌	-	❌
local_value		❌	-	❌
log_normal		✅	-	❌
log_sigmoid		❌	-	✅
log_sigmoid_forward		❌	-	❌
log_softmax	special_log_softmax	❌	-	✅
logaddexp		❌	-	❌
logaddexp2		❌	-	❌
logcumsumexp		❌	-	❌
logdet		❌	-	✅
logit	special_logit	❌	-	✅
logspace		❌	-	❌
logsumexp	special_logsumexp	❌	-	✅
lower		❌	-	❌
lstm		❌	-	✅
lstm_cell		❌	-	✅
lstrip		❌	-	❌
lstsq		❌	-	❌
lu_solve		❌	-	❌
lu_unpack		❌	-	❌
mH	adjoint	❌	-	❌
mT		❌	-	❌
manual_seed		❌	-	❌
margin_ranking_loss		❌	-	❌
masked_fill		✅	-	✅
masked_select		❌	-	✅
mathremainder		❌	-	❌
matmul	linalg_matmul	❌	-	✅
matrix_H		❌	-	❌
matrix_rank		❌	-	❌
max_pool1d		❌	-	✅
max_pool1d_with_indices		❌	-	✅
max_pool2d		❌	-	✅
max_pool3d		❌	-	✅
max_unpool2d		❌	-	❌
max_unpool3d		❌	-	❌
median		❌	-	❌
meshgrid		❌	-	✅
miopen_batch_norm		❌	-	❌
miopen_convolution		❌	-	❌
miopen_convolution_add_relu		❌	-	❌
miopen_convolution_relu		❌	-	❌
miopen_convolution_transpose		❌	-	❌
miopen_depthwise_convolution		❌	-	❌
miopen_rnn		❌	-	❌
mish		❌	-	✅
mkldnn_adaptive_avg_pool2d		❌	-	❌
mkldnn_convolution		❌	-	❌
mkldnn_linear		❌	-	❌
mkldnn_max_pool2d		❌	-	❌
mkldnn_max_pool3d		❌	-	❌
mkldnn_reorder_conv2d_weight		❌	-	❌
mkldnn_reorder_conv3d_weight		❌	-	❌
mkldnn_rnn_layer		❌	-	❌
mode		❌	-	❌
modf		❌	-	❌
movedim	moveaxis	❌	-	✅
mps_linear		❌	-	❌
mse_loss		❌	-	✅
msort		❌	-	❌
multi_margin_loss		❌	-	❌
multilabel_margin_loss		❌	-	❌
multilabel_margin_loss_forward		❌	-	❌
multinomial		❌	-	✅
mv		❌	-	✅
mvlgamma	special_multigammaln	❌	-	❌
nan_to_num		✅	-	✅
nanmean		❌	-	❌
nanmedian		❌	-	❌
nanquantile		❌	-	❌
nansum		❌	-	❌
narrow		❌	-	✅
narrow_copy		❌	-	❌
native_batch_norm		❌	-	❌
native_channel_shuffle		❌	-	❌
native_multi_head_self_attention		❌	-	❌
native_norm		❌	-	❌
neq		❌	-	❌
nested_to_padded_tensor		❌	-	❌
new_empty		❌	-	✅
new_empty_strided		❌	-	❌
new_full		❌	-	✅
new_ones		❌	-	✅
new_zeros		❌	-	✅
nextafter		✅	-	❌
nll_loss		❌	-	✅
nll_loss2d		❌	-	✅
nll_loss_forward		❌	-	❌
nll_loss_nd		❌	-	✅
node		❌	-	❌
nonzero_numpy		❌	-	✅
nonzero_static		❌	-	❌
norm		❌	-	✅
norm_except_dim		❌	-	❌
normal		✅	-	✅
normal_functional		❌	-	❌
nuclear_norm		❌	-	❌
numel		❌	-	✅
numpy_T		❌	-	✅
oct		❌	-	❌
one_hot		❌	-	✅
ones		❌	-	✅
ones_like		❌	-	✅
op		❌	-	❌
op_name		❌	-	❌
ord		❌	-	❌
ormqr		❌	-	❌
outer	ger	❌	-	✅
output_nr		❌	-	❌
owner		❌	-	❌
owner_name		❌	-	❌
pad		❌	-	✅
pad_sequence		❌	-	❌
pairwise_distance		❌	-	✅
partition		❌	-	❌
pdist		❌	-	❌
percentFormat		❌	-	❌
permute_copy		❌	-	❌
pin_memory		❌	-	❌
pinv		❌	-	❌
pinverse		❌	-	❌
pixel_shuffle		❌	-	✅
pixel_unshuffle		❌	-	✅
pointwise_placeholder		❌	-	❌
poisson		❌	-	❌
poisson_nll_loss		❌	-	❌
polar		❌	-	❌
polygamma	special_polygamma	✅	-	❌
pop		❌	-	✅
popitem		❌	-	❌
positive		❌	-	❌
prelu		❌	-	✅
promote_types		❌	-	❌
put		✅	-	❌
q_per_channel_axis		❌	-	❌
q_per_channel_scales		❌	-	❌
q_per_channel_zero_points		❌	-	❌
q_scale		❌	-	❌
q_zero_point		❌	-	❌
qr		❌	-	❌
qscheme		❌	-	❌
quantile		❌	-	❌
quantize		❌	-	❌
quantize_per_channel		❌	-	❌
quantize_per_tensor		❌	-	✅
quantize_per_tensor_dynamic		❌	-	❌
quantized_batch_norm		❌	-	❌
quantized_gru		❌	-	❌
quantized_lstm		❌	-	❌
quantized_max_pool1d		❌	-	❌
quantized_max_pool2d		❌	-	❌
quantized_max_pool3d		❌	-	❌
rad2deg		❌	-	❌
radians		❌	-	❌
rand_like		❌	-	✅
randint		❌	-	✅
randint_like		❌	-	✅
randn_like		❌	-	✅
random		✅	-	❌
range		❌	-	❌
ravel		❌	-	❌
real		❌	-	❌
record_stream		❌	-	❌
reduce_scatter_tensor		❌	-	❌
refine_names		❌	-	❌
relu6		✅	-	✅
remove		❌	-	❌
rename		✅	-	❌
renorm		❌	-	❌
repeat_interleave		❌	-	✅
replace		❌	-	❌
replication_pad1d		❌	-	✅
requires_grad_		❌	-	❌
reshape		❌	-	✅
reshape_as		❌	-	✅
resize		✅	-	❌
resize_as_		❌	-	❌
resize_as_sparse		✅	-	❌
resolve_conj		❌	-	✅
resolve_neg		❌	-	✅
result_type		❌	-	❌
retain_grad		❌	-	❌
retains_grad		❌	-	❌
reverse		❌	-	❌
rfind		❌	-	❌
rindex		❌	-	❌
rjust		❌	-	❌
rnn_relu		❌	-	✅
rnn_relu_cell		❌	-	❌
rnn_tanh		❌	-	✅
rnn_tanh_cell		❌	-	❌
roll		❌	-	✅
rot90		❌	-	❌
row_indices		❌	-	❌
row_indices_copy		❌	-	❌
rowwise_prune		❌	-	❌
rpartition		❌	-	❌
rrelu		❌	-	✅
rrelu_with_noise		✅	-	❌
rrelu_with_noise_functional		❌	-	❌
rsplit		❌	-	❌
rstrip		❌	-	❌
rsub		❌	-	✅
save		❌	-	❌
scaled_dot_product_attention		❌	-	✅
searchsorted		❌	-	❌
seed		❌	-	❌
segment_reduce		❌	-	❌
select_copy		❌	-	❌
selu		✅	-	✅
set		✅	-	❌
set_data		❌	-	❌
set_grad_enabled		❌	-	❌
set_source_Tensor_storage_offset		❌	-	❌
setdefault		❌	-	❌
sgn		❌	-	❌
signbit		❌	-	❌
silu		✅	-	✅
sinc	special_sinc	❌	-	❌
size		❌	-	✅
slice_copy		❌	-	❌
slice_inverse		❌	-	❌
slow_conv3d		❌	-	❌
slow_conv3d_forward		❌	-	❌
slow_conv_dilated2d		❌	-	❌
slow_conv_dilated3d		❌	-	❌
slow_conv_transpose2d		❌	-	❌
slow_conv_transpose3d		❌	-	❌
smm		❌	-	❌
smooth_l1_loss		❌	-	❌
soft_margin_loss		❌	-	❌
softmax	special_softmax	❌	-	✅
softplus		❌	-	✅
softshrink		❌	-	✅
solve		❌	-	❌
sorted		❌	-	❌
sparse_compressed_tensor		❌	-	❌
sparse_coo_tensor		❌	-	❌
sparse_dim		❌	-	❌
sparse_mask		❌	-	❌
sparse_resize		✅	-	❌
sparse_resize_and_clear		✅	-	❌
sparse_sampled_addmm		❌	-	❌
special_airy_ai		❌	-	❌
special_bessel_j0		❌	-	❌
special_bessel_j1		❌	-	❌
special_bessel_y0		❌	-	❌
special_bessel_y1		❌	-	❌
special_chebyshev_polynomial_t		❌	-	❌
special_chebyshev_polynomial_u		❌	-	❌
special_chebyshev_polynomial_v		❌	-	❌
special_chebyshev_polynomial_w		❌	-	❌
special_entr		❌	-	❌
special_erfcx		❌	-	❌
special_hermite_polynomial_h		❌	-	❌
special_hermite_polynomial_he		❌	-	❌
special_i0e		❌	-	❌
special_i1		❌	-	❌
special_i1e		❌	-	❌
special_laguerre_polynomial_l		❌	-	❌
special_legendre_polynomial_p		❌	-	❌
special_log_ndtr		❌	-	❌
special_modified_bessel_i0		❌	-	❌
special_modified_bessel_i1		❌	-	❌
special_modified_bessel_k0		❌	-	❌
special_modified_bessel_k1		❌	-	❌
special_ndtr		❌	-	❌
special_ndtri		❌	-	❌
special_scaled_modified_bessel_k0		❌	-	❌
special_scaled_modified_bessel_k1		❌	-	❌
special_shifted_chebyshev_polynomial_t		❌	-	❌
special_shifted_chebyshev_polynomial_u		❌	-	❌
special_shifted_chebyshev_polynomial_v		❌	-	❌
special_shifted_chebyshev_polynomial_w		❌	-	❌
special_spherical_bessel_j0		❌	-	❌
special_xlog1py		❌	-	❌
special_zeta		❌	-	❌
split		❌	-	✅
split_copy		❌	-	❌
split_with_sizes_copy		❌	-	❌
splitlines		❌	-	❌
square		✅	-	✅
sspaddmm		❌	-	❌
stack		❌	-	✅
startswith		❌	-	❌
std		❌	-	✅
std_mean		❌	-	✅
stft		❌	-	✅
storage_offset		❌	-	❌
str		❌	-	❌
stride		❌	-	❌
strip		❌	-	❌
sum_to		❌	-	❌
svd		❌	-	❌
swapcase		❌	-	❌
symbolic_b		❌	-	❌
symeig		❌	-	❌
t		✅	-	✅
take		❌	-	✅
take_along_dim		❌	-	❌
tanhshrink		❌	-	❌
tensor		❌	-	✅
tensor_split		❌	-	✅
tensordot		❌	-	✅
test		❌	-	❌
test_symbol		❌	-	❌
test_vartype		❌	-	❌
test_vartype2		❌	-	❌
thnn_conv2d		❌	-	❌
thnn_conv2d_forward		❌	-	❌
threshold		❌	-	✅
tile		❌	-	✅
title		❌	-	❌
to		❌	-	✅
to_dense		❌	-	❌
to_here		❌	-	❌
to_mkldnn		❌	-	❌
to_padded_tensor		❌	-	❌
trace		❌	-	❌
transpose	swapaxes, swapdims	✅	-	✅
transpose_copy		❌	-	❌
trapezoid		❌	-	❌
trapz		❌	-	❌
triangular_solve		❌	-	❌
tril		❌	-	✅
tril_indices		❌	-	❌
triplet_margin_loss		❌	-	❌
triu		❌	-	✅
triu_indices		❌	-	❌
type_as		❌	-	✅
unbind		❌	-	✅
unbind_copy		❌	-	❌
unflatten		❌	-	✅
unflatten_dense_tensors		❌	-	❌
unfold		❌	-	✅
unfold_copy		❌	-	❌
uniform		✅	-	❌
unique_consecutive		❌	-	❌
unique_dim		❌	-	✅
unique_dim_consecutive		❌	-	❌
unknown		❌	-	❌
unsafe_chunk		❌	-	✅
unsafe_split		❌	-	✅
unsafe_split_with_sizes		❌	-	✅
unsqueeze_copy		❌	-	❌
update		❌	-	❌
upper		❌	-	❌
upsample		❌	-	❌
upsample_bicubic2d		❌	-	✅
upsample_linear1d		❌	-	✅
upsample_nearest1d		❌	-	✅
upsample_nearest3d		❌	-	✅
upsample_trilinear3d		❌	-	✅
values		❌	-	❌
values_copy		❌	-	❌
vander		❌	-	❌
var_mean		❌	-	✅
vdot		❌	-	❌
view_as		❌	-	✅
view_as_complex		❌	-	❌
view_as_complex_copy		❌	-	❌
view_as_real		❌	-	❌
view_as_real_copy		❌	-	❌
view_copy		❌	-	❌
view_expand_placeholder		❌	-	❌
vsplit		❌	-	❌
vstack	row_stack	❌	-	✅
wait		❌	-	❌
wait_tensor		❌	-	❌
warn		❌	-	❌
warns		❌	-	❌
wrapped_linear_prepack		❌	-	❌
wrapped_quantized_linear_prepacked		❌	-	❌
xlogy	special_xlogy	❌	-	❌
zero		✅	-	✅
zeros		❌	-	✅
zeros_like		❌	-	✅
zfill		❌	-	❌