Repository navigation

All pull requests

Labels (19)Milestones (0)New pull request

Open

(40)

Closed

(501)

feat: Recognize model_connect as a TensorRT platform
#541yinggeh opened Oct 29, 2026Contributor7 of 20 tasks
3
docs: Correct inference response error ownership
#540autumn-wind opened Oct 26, 2026
1
fix: prevent crashes sorting equal-length cloud credentials
#539GeorgiaPreda opened Oct 26, 2026
1
perf: poll model repository models concurrently
#537drnextgis opened Oct 22, 2026
7
perf: avoid redundant S3 HeadBucket requests
#536drnextgis opened Oct 21, 2026
1
fix: Replace summary latency metrics with lock-free histograms
#535iis-MarkKuang opened Oct 20, 20262 tasks done
15
Fix rejected-request accounting and instance queue races
#532kozistr opened Oct 1, 2026
7
Fix dynamic batcher wake-up with finite max_queue_size
#531kozistr opened Oct 1, 2026
9
fix: Do not keep the model alive through Python API responses and requests
#529LiRunGuo opened Sep 24, 2026
build: build core in-tree against Conan-resolved dependencies
#524mc-nv opened Sep 8, 2026Contributor2 of 3 tasks
fix: remove distutils dependency
#519fwinkl opened Aug 5, 2026
2
Fix implicit sequence state output mapping
#506notandruu opened Jun 27, 2026
1
fix: validate ensemble composing model names against path traversal
#500LinZiyuu opened May 30, 2026
build: add Conan 2 + CMakePresets build system
#490mc-nv opened Apr 13, 2026Contributor
fix: Respect cc_model_filenames in Python backend auto-fill
#489nightflight-dk opened Apr 12, 2026Contributor
9
Validate that model outputs have non-empty names
#487mattrodriguez154 opened Apr 9, 2026
refactor: Centralized implementation for GetElementCount and GetByteSize
#485yinggeh opened Apr 7, 2026Contributor
Add GetByteSizeOverflow test for ValidateInputs overflow early-return path
#476yinggeh with CopilotAI opened Feb 26, 2026
feat: Add histogram latency metrics for all inference phases
#474zbennett10 opened Feb 24, 20266 tasks done
3
fix: Fix use-after-free in shared Metric objects during reload
#471aleksn7 opened Feb 16, 2026
Add TRITONSERVER_ServerOptionsSetLogCallback API
#463nbarbier-amira opened Dec 10, 2025
Add support for DefaultAzureCredential in Azure Storage backend
#451mkompanek opened Sep 16, 2025
fix: rate limiter deadlock due to head‑of‑line blocking
#448zeruniverse opened Aug 18, 2025
workflow tests check
#442yinggeh opened Jul 1, 2025Contributor
fix: Remove numpy<2 version constraint
#441yinggeh opened Jun 24, 2025Contributor
1

Navigation Menu

Repository navigation

All pull requests

feat: Recognize model_connect as a TensorRT platform

docs: Correct inference response error ownership

fix: prevent crashes sorting equal-length cloud credentials

perf: poll model repository models concurrently

perf: avoid redundant S3 HeadBucket requests

fix: Replace summary latency metrics with lock-free histograms

Fix rejected-request accounting and instance queue races

Fix dynamic batcher wake-up with finite max_queue_size

fix: Do not keep the model alive through Python API responses and requests

build: build core in-tree against Conan-resolved dependencies

fix: remove distutils dependency

Fix implicit sequence state output mapping

fix: validate ensemble composing model names against path traversal

build: add Conan 2 + CMakePresets build system

fix: Respect cc_model_filenames in Python backend auto-fill

Validate that model outputs have non-empty names

refactor: Centralized implementation for `GetElementCount` and `GetByteSize`

Add GetByteSizeOverflow test for ValidateInputs overflow early-return path

feat: Add histogram latency metrics for all inference phases

fix: Fix use-after-free in shared Metric objects during reload

Add TRITONSERVER_ServerOptionsSetLogCallback API

Add support for DefaultAzureCredential in Azure Storage backend

fix: rate limiter deadlock due to head‑of‑line blocking

workflow tests check

fix: Remove numpy<2 version constraint

Pull requests