v1.0.5

roboflow/inferencev1.0.5Mar 6, 2026by PawelPeczek-Roboflow

AI Summary

Adds model cold start response headers, supports YOLOLite in `inference_models`, and exposes health endpoints without an API key.

Key Highlights

  • Model cold start and ID response headers added
  • YOLOLite object detection support in `inference_models`
  • Exposes /healthz and /readiness endpoints without API key
  • Structured access logging gated by env var

New Features

  • Model cold start headers
  • YOLOLite support
  • Health endpoints without API key
  • Structured logging

Full Release Notes

## What's Changed
* feat: add model cold start, model/workflow/workspace ID response headers by @hansent in https://github.com/roboflow/inference/pull/2052
* Support yololite object detection in inference_models with ONNX backend by @leeclemnet in https://github.com/roboflow/inference/pull/2078
* Fix the input parameter types accepted by the Ethernet IP PLC block for PLC reads/writes by @shntu in https://github.com/roboflow/inference/pull/2061
* Try to address TRT issue by @PawelPeczek-Roboflow in https://github.com/roboflow/inference/pull/2079
* Release new inference-models by @PawelPeczek-Roboflow in https://github.com/roboflow/inference/pull/2084
* Email message serialization fix by @dkosowski87 in https://github.com/roboflow/inference/pull/2083
* Support New Roboflow API Usage Paused Error 423 by @maxschridde1494 in https://github.com/roboflow/inference/pull/2082
* Expose /healthz and /readiness endpoints even if API_KEY is not set by @ecarrara in https://github.com/roboflow/inference/pull/2077
* feat: inference_models adapters respect countinference for credit verification bypass by @hansent in https://github.com/roboflow/inference/pull/2081
* Update docs by @Erol444 in https://github.com/roboflow/inference/pull/2076
* Reduce flash-attn MAX_JOBS to 1 for JP7.1 build by @alexnorell in https://github.com/roboflow/inference/pull/2068
* Bump the npm_and_yarn group across 2 directories with 10 updates by @dependabot[bot] in https://github.com/roboflow/inference/pull/2085
* Fix shared model cache race conditions causing pod crashes by @hansent in https://github.com/roboflow/inference/pull/2080
* fix inference-models pypi publishing by @grzegorz-roboflow in https://github.com/roboflow/inference/pull/2086
* Qwen3 5 and move to transformers 5 by @Matvezy in https://github.com/roboflow/inference/pull/2070
* Correct resize procedure for RF-DETR models trained on versions with non-stretch, non-square resize by @mkaic in https://github.com/roboflow/inference/pull/2067
* ENT-969: Add TestPatternStreamProducer as a built-in video source type by @NVergunst-ROBO in https://github.com/roboflow/inference/pull/2056
* fix: handle expired Redis lock release gracefully by @rafel-roboflow in https://github.com/roboflow/inference/pull/2060
* Revert/qwen 3.5 by @PawelPeczek-Roboflow in https://github.com/roboflow/inference/pull/2087
* feat: gate structured access logging behind STRUCTURED_API_LOGGING env var by @hansent in https://github.com/roboflow/inference/pull/2088
* Deploy inference-models-0.19.5 by @PawelPeczek-Roboflow in https://github.com/roboflow/inference/pull/2089

## New Contributors
* @maxschridde1494 made their first contribution in https://github.com/roboflow/inference/pull/2082
* @ecarrara made their first contribution in https://github.com/roboflow/inference/pull/2077

**Full Changelog**: https://github.com/roboflow/inference/compare/v1.0.4...v1.0.5