ultralytics 8.4.38

v8.4.38 - `ultralytics 8.4.38` Unify args naming for standalone export functions (#24120)

on Python PyPI

🌟 Summary

Ultralytics v8.4.38 is a stability-focused release that mainly improves model export reliability and consistency across many deployment formats, with additional fixes for training, tracking, and SAM3 behavior. 🚀

📊 Key Changes

• Top priority (PR #24120 by @Laughing-q): Unified standalone export behavior across CoreML, ONNX, OpenVINO, TensorFlow, TorchScript, ExecuTorch, Axelera, RKNN, IMX, NCNN, MNN, and Paddle. 📦
  • CoreML now uses the model’s real input name (instead of assuming "image"), improving compatibility.
  • Export helper functions were standardized with clearer argument names (output_file, output_dir, etc.).
  • Better support for multi-input export scenarios (notably ONNX/OpenVINO).
  • More robust path handling in TensorFlow/IMX export flows.
• Export robustness fixes beyond the main PR
  • Fixed YOLOE export edge case by skipping incompatible fusion when lrpc is present.
  • Improved dependency auto-install fallback to use python -m pip (works better in restricted environments).
  • RKNN now requires newer supported toolkit versions (rknn-toolkit2>=2.3.2).
  • Axelera install commands now allow prerelease packages (--pre) where needed.
• Training reliability improvements 🏋️
  • Fixed resume for non-end2end models (like classification).
  • Adjusted DDP setup order so stride/image-size checks happen safely before wrapping.
  • Checkpoint saves are now skipped whenever EMA weights contain invalid values (NaN/Inf), preventing bad checkpoints.
• Tracking behavior correction 🎯
  • track_buffer now consistently behaves as a true frame count in ByteTrack/BoT-SORT (no hidden FPS scaling).
• SAM3 fixes 🧠
  • Presence-logit clamp now actually applies (in-place).
  • Text-only prompting no longer injects a dummy geometry token, improving confidence behavior.

🎯 Purpose & Impact

• For deployment users: exports should be more predictable and less fragile across devices and runtimes (especially CoreML/OpenVINO/ONNX pipelines). ✅
• For training users: fewer resume/DDP/checkpoint edge-case failures, especially in classification and multi-GPU setups. ⚙️
• For tracking users: track_buffer now matches what you configure, reducing surprising track drop behavior at non-30 FPS. 🎥
• For SAM3 users: improved numerical stability and better text-only grounding quality. 🔬
• Overall: this release is less about new features and more about making existing workflows safer, clearer, and production-friendlier. 🙌

What's Changed

• docs: add trackzone to solutions cli enum by @raimbekovm in #24194
• Fix docs macros type and default drift by @raimbekovm in #24195
• Fix train resume for non-end2end models by @fcakyon in #24173
• Fix torch and torchvision versions on JetPack 5 Docker by @lakshanthad in #24199
• fix: update axelera-devkit installation command to allow prerelease versions by @onuralpszr in #24190
• Refactor test_engine.py and add OBB/Pose test coverage by @Laughing-q in #24197
• Fix docs drift in export-table for executorch, axelera, imx by @raimbekovm in #24189
• Fix solutions-args colormap default drift (JET vs DEEPGREEN) by @raimbekovm in #24209
• Fix trainer.stride assignment for DDP training by @Laughing-q in #24208
• Add missing validated args to TorchScript, CoreML, ExecuTorch integration pages by @raimbekovm in #24207
• Fix SAM3 presence logits clamp being a no-op by @Y-T-G in #24213
• Clarify YOLO12 pretrained weights are detect-only in model docs by @raimbekovm in #24211
• docs: remove outdated Snyk badge from security documentation and update old links by @onuralpszr in #24221
• Update annotation viewer zoom shortcut docs by @laodouya in #24214
• Fix Sentry bugs 2026-04-14 by @glenn-jocher in #24220
• Fix table format in annotation.md by @Laughing-q in #24232
• Update axelera-runtime installation command to allow prerelease versions by @lakshanthad in #24230
• fix: OpenVINO 2026 segfault on conda CI by @onuralpszr in #24224
• Clarify tt100k uses 221 annotation categories with 45 trainable classes by @raimbekovm in #24234
• Docs: add cli tabs for all 12 yolo solutions by @raimbekovm in #24217
• Fix: corrected undocumented scaling of track_buffer in trackers by @TimSchoonbeek in #24247
• Check for lrpc attribute before attempting fuse during YOLOE model export by @Y-T-G in #24239
• Return the model instance when calling fuse method by @Y-T-G in #24246
• SAM3: Skip geometry token when using text prompts by @Y-T-G in #24244
• Update solution tests to use cached session assets by @Laughing-q in #24237
• ultralytics 8.4.38 Unify args naming for standalone export functions by @Laughing-q in #24120

New Contributors

• @TimSchoonbeek made their first contribution in #24247

Full Changelog: v8.4.37...v8.4.38