b1tg 45e2f916a3 add quantize fp8 in llama3 (#12893) ...

* add quantize fp8 in llama3

* don't truncate fp8 alu result

* cast to float32 before matmul

* --model weights/LLaMA-3/8B-SF-DPO/

---------

Co-authored-by: chenyu <chenyu@fastmail.com>

2025-10-27 10:22:57 -04:00

..

conversation_data

Whisper + LLAMA + VITS (#2332)

2023-12-02 15:03:46 -08:00

llm.c

clean up unused imports in examples and update CI linting (#11024)

2025-06-30 08:21:27 -07:00

mlperf

BS=96 for bert (#12675)

2025-10-14 09:07:43 -04:00

openpilot

compile3: make selftests opt-in (#12851)

2025-10-21 11:32:27 -07:00

other_mnist

move frontend dir to nn [pr] (#12470)

2025-10-07 10:42:22 +08:00

rl

more beautiful_cartpole with exposed hparams

2024-01-07 17:41:09 -08:00

sovits_helpers

combine pad2d with pad (#7677)

2024-11-14 17:56:02 +08:00

tinychat

rename lazydata to uop (#10698)

2025-06-08 08:42:22 -07:00

vgg7_helpers

leakyrelu to leaky_relu (#9270)

2025-02-26 13:22:08 -05:00

webgpu

rename lazydata to uop (#10698)

2025-06-08 08:42:22 -07:00

__init__.py

failing llama test

2023-03-11 16:28:10 -08:00

beautiful_cartpole.py

remove Tensor.no_grad, it's meaningless now [pr] (#10556)

2025-05-28 22:20:02 -07:00

beautiful_cifar.py

remove FUSE_ARANGE (#12511)

2025-10-08 04:54:07 -04:00

beautiful_mnist_multigpu.py

Fix mypy examples/beautiful_*.py (#6978)

2024-10-10 11:34:29 -04:00

beautiful_mnist.py

test backward in test_tiny (#11697)

2025-08-16 20:29:39 -07:00

benchmark_onnx.py

move frontend dir to nn [pr] (#12470)

2025-10-07 10:42:22 +08:00

coder.py

clean up unused imports in examples and update CI linting (#11024)

2025-06-30 08:21:27 -07:00

compile_efficientnet.py

CLANG -> CPU (#9189)

2025-02-20 18:03:09 -05:00

compile_tensorflow.py

move frontend dir to nn [pr] (#12470)

2025-10-07 10:42:22 +08:00

conversation.py

remove Tensor.no_grad, it's meaningless now [pr] (#10556)

2025-05-28 22:20:02 -07:00

efficientnet.py

remove clang program header (#4422)

2024-05-04 08:38:01 -07:00

flux1_seed0.png

Flux.1 (#6334)

2024-09-24 10:08:04 +08:00

flux1.py

flux set model path in args (#7660)

2024-11-12 22:11:40 -05:00

gpt2.py

fix gpt2 with benchmark (#12736)

2025-10-16 09:55:20 -04:00

hlb_cifar10.py

remove FUSE_ARANGE (#12511)

2025-10-08 04:54:07 -04:00

llama3.py

add quantize fp8 in llama3 (#12893)

2025-10-27 10:22:57 -04:00

llama.py

replace hardcoded GPU in llama debug msg (#12102)

2025-09-10 13:56:40 -04:00

mamba.py

added top k sampling to examples/mamba (#12061)

2025-09-14 15:27:34 -04:00

mask_rcnn.py

change Tensor.stack to method (#4719)

2024-05-24 17:04:19 -04:00

minrf.py

remove Tensor.no_grad, it's meaningless now [pr] (#10556)

2025-05-28 22:20:02 -07:00

mixtral.py

Subtract 1 from Variable upper bound (#10715)

2025-06-09 09:25:53 -07:00

mnist_gan.py

leakyrelu to leaky_relu (#9270)

2025-02-26 13:22:08 -05:00

olmoe.py

remove .float calls in olmoe (#11610)

2025-08-10 20:33:22 -04:00

openelm.py

nn.RMSNorm (#5272)

2024-07-02 21:39:01 -04:00

qwq.py

replace hardcoded GPU in llama debug msg (#12102)

2025-09-10 13:56:40 -04:00

sdv2.py

stable diffusion --fakeweights (#12810)

2025-10-20 12:41:06 +02:00

sdxl_seed0.png

fix failed threefry (#10646)

2025-06-05 17:17:42 -07:00

sdxl.py

don't validate output in sdxl with fakeweights (#12160)

2025-09-13 21:47:51 -04:00

self_tokenize.py

make self_tokenize output more like a python file (#8411)

2024-12-25 14:16:30 -05:00

serious_mnist.py

combine pad2d with pad (#7677)

2024-11-14 17:56:02 +08:00

simple_conv_bn.py

fix various examples (#4691)

2024-05-22 20:43:21 -04:00

so_vits_svc.py

small fix replacing download_file with fetch (#10877)

2025-06-19 12:12:09 -04:00

stable_diffusion_seed0.png

default threefry (#6116)

2024-09-25 17:45:13 +08:00

stable_diffusion.py

stable diffusion --fakeweights (#12810)

2025-10-20 12:41:06 +02:00

stunning_mnist.py

clean up unused imports in examples and update CI linting (#11024)

2025-06-30 08:21:27 -07:00

test_onnx_imagenet.py

delete DONT_REALIZE_EXPAND and DONT_GROUP_REDUCES (#12744)

2025-10-16 14:11:33 -04:00

test_pkl_imagenet.py

more stuff from DSP (#9689)

2025-04-02 15:27:48 +08:00

torch_cuda_kernel.py

clean up unused imports in examples and update CI linting (#11024)

2025-06-30 08:21:27 -07:00

train_efficientnet.py

tinytqdm.set_description and tinytrange (#5101)

2024-06-22 14:45:06 -04:00

train_resnet.py

fix fromarray depreceation (#12512)

2025-10-08 09:13:26 -04:00

transformer.py

fixing transformer training bug (#9877)

2025-04-13 19:34:20 -04:00

vgg7.py

clean up unused imports in examples and update CI linting (#11024)

2025-06-30 08:21:27 -07:00

vit.py

move to new cached fetch (#2493)

2023-11-28 17:36:55 -08:00

vits.py

update vits vctk model to use download from huggingface (#10688)

2025-06-07 20:47:28 -04:00

whisper.py

Tensor.pad_to and Tensor.shrink_to (#12210)

2025-09-16 12:24:55 -04:00

yolov3.py

fix bugs at examples/yolov3.py (#11614)

2025-08-11 21:14:47 -04:00

yolov8-onnx.py

move frontend dir to nn [pr] (#12470)

2025-10-07 10:42:22 +08:00

yolov8.py

clean up unused imports in examples and update CI linting (#11024)

2025-06-30 08:21:27 -07:00