Issues

• Find a less error-prone method to implement version pinning for RAFT and RMM

  triton-inference-server/fil_backend#464 · Filter by author chyunsu3 opened on Sep 15, 2026

• 25.09 build broken, maintainers open to cherry pick fixes?

  triton-inference-server/fil_backend#448 · Filter by author a-ys opened on Feb 14, 2026
  1

• Get raw Booster prediction instead of model.predict_proba values from an XGBoost model

  triton-inference-server/fil_backend#439 · Filter by author xlr18 opened on Sep 22, 2025
  1

• Legacy FIL produces incorrect output for latest LightGBM

  triton-inference-server/fil_backend#432 · Filter by author chyunsu3 opened on May 21, 2025

• Does Nvidia Triton Inference Server Support AutoML(AutoGluon) framework?

  triton-inference-server/fil_backend#410 · Filter by author IamExperimenting opened on Oct 21, 2024

• Does the FIL backend support the XGBoost base_margin feature?

  enhancement
  triton-inference-server/fil_backend#408 · Filter by author leslizhang opened on Oct 13, 2024
  1

• [CI] Collect dependency versions in a single YAML file

  triton-inference-server/fil_backend#407 · Filter by author chyunsu3 opened on Oct 2, 2024

• Error while running triton-server docker with fil backed on MacOS 14.3

  triton-inference-server/fil_backend#400 · Filter by author suvratjain1995 opened on Aug 21, 2024
  1

• [FEA] Support categorical features when serving XGBoost models

  enhancement
  triton-inference-server/fil_backend#389 · Filter by author gfalcone opened on May 15, 2024
  2

• How to limit the number of CPU cores used by fil backend？

  triton-inference-server/fil_backend#385 · Filter by author SkyM31 opened on Apr 30, 2024

• LLVM ERROR: out of memory

  triton-inference-server/fil_backend#368 · Filter by author sandeepb2013 opened on Oct 30, 2023
  18

• Provide ARM release to allow support for FIL backend on Jetson and other ARM platforms

  triton-inference-server/fil_backend#362 · Filter by author blthayer opened on Jul 12, 2023
  20

• [BUG] cuml binary classification models do not observe threshold

  triton-inference-server/fil_backend#351 · Filter by author RAMitchell opened on Mar 28, 2023
  1

• [BUG] Multiclass models must have output_class=true to predict probabilities

  triton-inference-server/fil_backend#350 · Filter by author RAMitchell opened on Mar 28, 2023

• [FEA] Provide CatBoost Support

  enhancement
  triton-inference-server/fil_backend#347 · Filter by author riaris opened on Mar 23, 2023
  4

• [Bug] Wrong output shapes for Shapley values

  triton-inference-server/fil_backend#342 · Filter by author RAMitchell opened on Mar 10, 2023
  3

• [DOC] More clearly document that max_batch_size 0 is not supported

  triton-inference-server/fil_backend#338 · Filter by author wphicks opened on Feb 10, 2023

• [FEA] Support max_batch_size 0

  triton-inference-server/fil_backend#337 · Filter by author wphicks opened on Feb 10, 2023

• [BUG] Segfault for max_batch_size 0

  triton-inference-server/fil_backend#336 · Filter by author wphicks opened on Feb 10, 2023

• Server segfault if output dims missing from config.pbtxt

  triton-inference-server/fil_backend#330 · Filter by author wphicks opened on Jan 30, 2023

• Wrong error trace when LightGBM model file is not found

  triton-inference-server/fil_backend#328 · Filter by author jliebana opened on Jan 20, 2023
  1

• Internal Docs Feedback: Provide a notebook demonstrating deployment of a pre-trained model

  triton-inference-server/fil_backend#295 · Filter by author wphicks opened on Oct 3, 2022

• Internal Docs Feedback: Provide sample model in NGC container

  triton-inference-server/fil_backend#293 · Filter by author wphicks opened on Oct 3, 2022

• Does fil support lightgbm predict leaf index?

  triton-inference-server/fil_backend#273 · Filter by author chenglin opened on Jul 14, 2022
  3

• Add example of submitting a Python request with shared memory to FAQ notebook

  triton-inference-server/fil_backend#270 · Filter by author wphicks opened on Jun 30, 2022