Search results

Effective MLComputePolicy exposure

policy selection

Status: Open.

#934 In webmachinelearning/webnn;

· anssiko opened on May 7, 2026

Define an upper bound on number of concat inputs and split outputs

security-trackerGroup bringing to attention of security, or tracked by the security Group but not needing response.Group bringing to attention of security, or tracked by the security Group but not needing response.

Status: Open.

#931 In webmachinelearning/webnn;

· philloooo opened on Apr 24, 2026

RFE: Add support for more floating point low-precision ML data types (bfloat16, fp8, nvfp4)

feature request

opset

Status: Open.

#930 In webmachinelearning/webnn;

· mtavenrath opened on Apr 21, 2026

RFE: Add batching support (N-D tensors) to the gemm operator

Status: Open.

#929 In webmachinelearning/webnn;

· mtavenrath opened on Apr 20, 2026

Consider specify lower limit for conv2d/pool2d kernel sizes, dilations, strides

security-trackerGroup bringing to attention of security, or tracked by the security Group but not needing response.Group bringing to attention of security, or tracked by the security Group but not needing response.

Status: Open.

#928 In webmachinelearning/webnn;

· philloooo opened on Apr 2, 2026

Security considerations refresh

security-trackerGroup bringing to attention of security, or tracked by the security Group but not needing response.Group bringing to attention of security, or tracked by the security Group but not needing response.

Status: Open.

#927 In webmachinelearning/webnn;

· anssiko opened on Mar 27, 2026

Mandate output_channels % groups == 0 validation for conv2d

security-trackerGroup bringing to attention of security, or tracked by the security Group but not needing response.Group bringing to attention of security, or tracked by the security Group but not needing response.

Status: Open.

#925 In webmachinelearning/webnn;

· lynne0326 opened on Mar 20, 2026

Consider specifying tensor element count limit

feature request

Status: Open.

#924 In webmachinelearning/webnn;

· philloooo opened on Mar 19, 2026

Creating a context should be done using finer-grained in-parallel steps

conventions

Status: Open.

#919 In webmachinelearning/webnn;

· gterzian opened on Feb 17, 2026

Floating point accuracy for sin and cos with range constraint

operator specific

Status: Open.

#914 In webmachinelearning/webnn;

· philloooo opened on Jan 5, 2026

accelerated should be prior to powerPreference for device selection

policy selection

Status: Open.

#911 In webmachinelearning/webnn;

· mingmingtasd opened on Dec 18, 2025

Composite operators / subgraphs

opset

Status: Open.

#907 In webmachinelearning/webnn;

· fdwr opened on Nov 27, 2025