Search results

./test/float8/test_everything_multi_gpu.sh failing in 4xH100 CI

float8

module: trainingquantize_ api training flowquantize_ api training flow

Status: Open.

Bug

#4106 In pytorch/ao;

· danielvegamyhre opened on Mar 18, 2026

·  

FP8 Rowwise Training

Blockwise FP8 training seems to have much lower peak memory usage than others

float8

module: trainingquantize_ api training flowquantize_ api training flow

Status: Open.

#4101 In pytorch/ao;

· CFC87 opened on Mar 17, 2026

·  

FP8 Blockwise Training

FP8 Blockwise Training Tracker

float8

good first issueGood for newcomersGood for newcomers

module: trainingquantize_ api training flowquantize_ api training flow

moe

tracker

Status: Open.

#3290 In pytorch/ao;

· danielvegamyhre opened on Nov 4, 2025

·  

FP8 Blockwise Training

question of blockwise quant fp8 training

float8

moe

Status: Open.

Feature

#3226 In pytorch/ao;

· rakkit opened on Oct 22, 2025

FLUX FP8 Training Error

bugSomething isn't workingSomething isn't working

float8

Status: Open.

Bug

#3111 In pytorch/ao;

· faruknane opened on Oct 1, 2025

Gradient Checkpoint makes FP8 Training Slow

float8

Status: Open.

#2445 In pytorch/ao;

· Yzx835 opened on Jun 26, 2025

[float8] Investigate if workaround for slow cutlass rowwise GEMM when fast_accum=False is still needed after perf improvments and potentially optimize GEMM further

float8

Status: Open.

#2184 In pytorch/ao;

· danielvegamyhre opened on May 8, 2025

[float8] Support power of 2 scales with PerRow scales for inference

float8

Status: Open.

#2182 In pytorch/ao;

· danielvegamyhre opened on May 7, 2025

cast to mxfp8 across dim1 should be performant

float8

triaged

Status: Open.

#1945 In pytorch/ao;

· vkuzo opened on Mar 24, 2025

[Bug] FSDP2 FP8 compatibility problem with nn.Linear layers (GPU count > out_features)

distributed

float8

triaged

Status: Open.

#1938 In pytorch/ao;

· HIT-cwh opened on Mar 24, 2025

fp8 quantization with FSDP2 error

float8

quantize_quantize_ APIquantize_ API

triaged

Status: Open.

#1929 In pytorch/ao;

· happynear opened on Mar 20, 2025

Does torchao support FP8 Grouped GEMM?

float8

triaged

Status: Open.

#1928 In pytorch/ao;

· zigzagcai opened on Mar 20, 2025