Skip to content

A WebGL accelerated JavaScript library for training and deploying ML models.

active 2024-12-302026-08-05 (UTC)

Partial coverage12,420 / 14,129 hourly files (88%) · 2 absent upstream · 1,703 failed, retryable2024-12-302026-08-11 (UTC)— sampled evenly across the window, so rankings and trends hold; absolute counts scale up.
Events
1K
Pushes
4
Pull requests
71
Issues
183
Stars
374
Forks
52

Activity over time

Daily event counts in the loaded window

Line chart, 584 days from 2024-12-30 to 2026-08-05. Pushes: 4 total, peak 2 in a day. Pull requests: 71 total, peak 10 in a day. Issues: 183 total, peak 8 in a day. Comments: 256 total, peak 7 in a day. Stars: 374 total, peak 6 in a day.

  • Pushes
  • Pull requests
  • Issues
  • Comments
  • Stars

Stars, PRs, issues and forks are under-captured in the later part of this window. GH Archive progressively stopped capturing non-push events during 2026 — −95% or worse by the end of the window. Every series here except Pushes fades for that reason, so a decline above reflects the archive, not this repository. Pushes stay reliable throughout, so read them, and the contributor counts derived from them, as the real signal. Data health has the measurements.

Top contributors

Pushes, PRs, issues, reviews and comments — stars and forks excluded, so this is contribution rather than popularity

Recent activity

Latest issues, pull requests and releases

  • Issue comment#8393Varun-S102026-08-05 04:53
    The output of tf.depthwiseConv2d operation on the WASM backend is inconsistent with the output on the CPU and WebGL backends
  • Issue comment#7271github-actions[bot]2026-07-31 02:18
    Infinite loop in tf.loadLayersModel
  • Issue#8471Varun-S102026-07-22 05:29
    Shared layer weights are leaked when disposing a model
  • Issue#5449github-actions[bot]2026-07-16 02:18
    ' ReferenceError: FormData is not defined ' when attempting to call a POST endpoint to save a model (tfjs-node -> tfjs-core))
  • Issue#5381github-actions[bot]2026-07-15 02:17
    layersModel working for inference but breaks when training
  • Issue comment#5490github-actions[bot]2026-07-15 02:17
    Blazeface result is not correct on WASM backend
  • Issue comment#7961github-actions[bot]2026-07-15 02:17
    tfvis.metrics.confusionMatrix leaks tensors
  • Issue#7961github-actions[bot]2026-07-15 02:17
    tfvis.metrics.confusionMatrix leaks tensors
  • Issue#8733alanvww2026-07-04 20:21
    [webgpu] Video frames imported via importExternalTexture are misoriented on iOS 26 (landmarks rotated ~90°)
  • Issue comment#8628google-ml-butler[bot]2026-07-04 02:19
    Path Traversal Vulnerability in tfjs-converter write_weights.py Allows Arbitrary File Write
  • Issue#8569github-actions[bot]2026-07-04 02:19
    Bug: Nested tf.layers.Layer fails to register weights (Total params: 0) despite a valid WebGL context
  • Issue comment#8569github-actions[bot]2026-07-04 02:19
    Bug: Nested tf.layers.Layer fails to register weights (Total params: 0) despite a valid WebGL context
  • Issue#8635github-actions[bot]2026-07-04 02:19
    [Converter] FailedPreconditionError when output shard path is a directory (Missing path validation)
  • Issue comment#8635github-actions[bot]2026-07-04 02:19
    [Converter] FailedPreconditionError when output shard path is a directory (Missing path validation)
  • Issue comment#7722google-ml-butler[bot]2026-07-02 02:21
    Help! Does this work at all?
  • Issue comment#7617google-ml-butler[bot]2026-06-28 02:22
    requested allocation size 0xd55559f0 exceeds maximum supported size of 0xc0000000 when running tf.conv2d on wasm backend
  • Issue comment#7612github-actions[bot]2026-06-28 02:22
    A null-pointer-dereference error is reported on the tf.layers.conv2d function when using the WebAssembly (WASM) backend
  • Issue comment#7621google-ml-butler[bot]2026-06-28 02:22
    location-size-too-big (tfjs-backend-wasm.wasm+0x51ae8d) in malloc+0x51ae8d when running tf.conv2d on wasm backend
  • Issue comment#7461google-ml-butler[bot]2026-06-27 02:21
    pow(1, NaN) differs between cpu and wasm backend
  • Issue comment#8325google-ml-butler[bot]2026-06-27 02:21
    Input layer's input shape is changed by tf.layers.centerCrop height and width argument.
  • Issue#7461github-actions[bot]2026-06-27 02:21
    pow(1, NaN) differs between cpu and wasm backend
  • Issue comment#8395google-ml-butler[bot]2026-06-27 02:21
    The output of the tf.dot operation on the CPU backend is inconsistent with the output on the WebGL, Wasm, and TensorFlow backends
  • Issue#7914github-actions[bot]2026-06-27 02:21
    WebGPU backend not found in webworkers
  • Issue comment#8373arcman72026-06-26 05:22
    Dispatch size exceeds WebGPU limits in Y or Z dimension
  • Issue comment#8322google-ml-butler[bot]2026-06-26 02:22
    The argument numTokens will impact inputLayer's output shape

Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 374 stars here means stars gained during the window, not the repo's star count.