v0.64.8

💪 Added

• Fisheye cameras in camera calibration block by @Erol444 in #1996
  Calibration block was supporting polynomial calibration which is not handling fisheye distortions well. This change adds support for fisheye calibration.

• Heatmap block by @Erol444 in #1986
  This change adds heatmap block (uses supervision's heatmap annotator), which supports both:

• detections, so heatmap based on where detections were
• tracklets, which ignores stationary objects (default: on), so we heatmap the movements not the objects

heatmap2.mp4

🚧 Maintanence

• temporarily pin z3-solver version by @grzegorz-roboflow in #1990
• Code workflow block icon issue by @Erol444 in #1988
• Optimize cosine_similarity by @KRRT7 in #1989
• add inference version to the request headers by @japrescott in #1985
• Fix video frame count estimation by detecting actual FPS from uploaded video by @rafel-roboflow in #1992
• Mark file processing in webrtc worker for downstream blocks to pick frame timestamp correctly by @grzegorz-roboflow in #1995
• add frame size to webrtc video metadata by @rafel-roboflow in #1997
• enable gzip compression by default by @rafel-roboflow in #1998
• WIP: enabled sam3 visual segment by @rafel-roboflow in #1975
• added ffmpeg to docker dependencies by @rafel-roboflow in #2002
• rename seg-preview to sam3 by @rafel-roboflow in #2005
• Fix RF-DETR-Seg mask postprocessing for letterboxed input case by @mkaic in #2001
• Enable inference pipeline api on jetpack 6.2.0 by @grzegorz-roboflow in #2006

Full Changelog: v0.64.7...v0.64.8

v1.0.0rc1

inference 1.0.0rc1 — Release Candidate

Today marks an important milestone for Inference.

Over the past years, Inference has grown from a lightweight prediction server into a widely adopted runtime used across local deployments, Docker, edge devices, and production systems. Hundreds of releases later, the project has matured significantly — and so has the need for a faster, more modular, and future-proof.

inference 1.0.0rc1 is a preview of 1.0.0 release which will close one chapter and open another - this release introduces a new prediction engine that will become the foundation for all future development.

🚀 New prediction engine - inference-models

We are introducing inference-models, a redesigned execution engine focused on:

• faster model loading and inference
• improved resource utilization
• better modularity and extensibility
• cleaner separation between serving and model runtime
• stronger foundations for future major versions

The engine is already available today in:

• inference-models package → 0.18.6rc8 (RC)
• inference package and Docker → enabled with env variable

USE_INFERENCE_MODELS=True

inference-models wrapped within old inference is a drop-down replacement. This allows testing the new runtime without changing existing integrations.

Important

Predictions from your models may change - but generally for better! inference-models is completely new engine for running models, we have fixed a lot of bugs and make it multi-backend - capable to run onnx, torch and even trt models! It automatically negotiate with Roboflow model registry to choose best package to run in your environment. We have already migrated almost all Roboflow models to new registry - working hard to achieve full coverage soon!

📅 What happens next

• Next week

  • Stable Inference 1.0.0
  • Stable inference-models release
  • Roboflow platform updated to use inference-models as the default engine

• In the coming weeks

  • inference-models becomes the default engine for public builds (USE_INFERENCE_MODELS becomes opt-out, not opt-in)
  • continued performance improvements and runtime optimizations

🔭 Looking forward - the road to 2.0

• This engine refresh is only the first step.
• We are starting work toward Inference 2.0, a larger modernization effort similar in spirit to the changes introduced with inference-models.

Stay tuned for future updates!

v0.64.7

What's Changed

• dg 15 fix timeout file by @rafel-roboflow in #1934
• Fix VLM as Detector/Classifier name, so it gets correct URL by @Erol444 in #1965
• improve error logging by @japrescott in #1966
• Add rfdetr nas by @probicheaux in #1970
• added missing envvar export for webrtc preview gzip flag by @rafel-roboflow in #1978
• Bug/dg 204 (2) reduce ack window webrtc by @rafel-roboflow in #1979
• Claude opus 4.6 in Claude block by @Erol444 in #1980
• Add remote exec capability for foundation models missing it by @hansent in #1968
• Gemini block: Add support for tool code execution (tool use) by @Erol444 in #1961
• Pass delete from disk to clear cache by @bigbitbus in #1982
• Add change to avoid pushing latest tag for rc release by @PawelPeczek-Roboflow in #1983

Full Changelog: v0.64.6...v0.64.7

v0.64.6

What's Changed

• Add large rf-detrs and seg coco models to inference_models by @Matvezy in #1944
• Add configurable RF API timeout for inference-cli command interacting with RF-cloud by @PawelPeczek-Roboflow in #1950
• Allow sv.Detections.data properties in extract-property block by @grzegorz-roboflow in #1948
• CI for deploying custom python block modal app by @grzegorz-roboflow in #1945
• in modal.custom_python_block.deploy.yml use deployment modal tokens by @grzegorz-roboflow in #1953
• Address internals imported from Supervision by @grzegorz-roboflow in #1951
• Add YOLO26 to inference_models by @mkaic in #1943
• Fix failing yolo26 gpu integration tests by @mkaic in #1956
• Disable automatic deployment of modal webexec by @grzegorz-roboflow in #1954
• 0.64.6 by @grzegorz-roboflow in #1957

Full Changelog: v0.64.5...v0.64.6

v0.64.5

What's Changed

• Fix 2 urls by @Erol444 in #1939
• Reduce number of layers in lambda dockerfile by @grzegorz-roboflow in #1940
• Add xl rfdetrs types by @Matvezy in #1937
• Add collimate algorithm and Otsu improvements to OCR stitch block by @reiffd7 in #1936
• Fix SAM3 code sample by @Erol444 in #1941
• Feature/dg 3 add preview flag to video streaming previews by @rafel-roboflow in #1942

New Contributors

• @Erol444 made their first contribution in #1939

Full Changelog: v0.64.4...v0.64.5

v0.64.4

What's Changed

• Address numpy.fromstring deprecation by @grzegorz-roboflow in #1925
• fix documentation issue with special chars by @rafel-roboflow in #1926
• Fix/dg 18 rotation metadata for videos is not applied serverside by @rafel-roboflow in #1911
• First iteration of inference-models docs by @PawelPeczek-Roboflow in #1922
• Action install ffmpeg modal by @rafel-roboflow in #1928
• added search keywords metadata to image preprocessing block by @rafel-roboflow in #1931
• Cache describe_interface and other endpoints by @yeldarby in #1932
• Add rfdetr large, xl, xxl, and seg aliases by @Matvezy in #1933
• Safeq qwen test by @Matvezy in #1900
• Enable crash dump of input image by @grzegorz-roboflow in #1921
• Feature/event log block improvements by @jeku46 in #1929
• Selectively disabling workflow blocks from inference server by @bigbitbus in #1924

Full Changelog: v0.64.3...v0.64.4

v0.64.3

What's Changed

• Fix yolo26 XL Seg aliases by @Matvezy in #1919
• Fix rfdetr instance segmentation and object detection postprocessing error by @grzegorz-roboflow in #1920
• Add docs about YOLO26 by @PawelPeczek-Roboflow in #1923

Full Changelog: v0.64.2...v0.64.3

v0.64.2

What's Changed

• Add more yolo26 aliases by @Matvezy in #1918

Full Changelog: v0.64.1...v0.64.2

v0.64.1

What's Changed

• Add yolo26 aliases by @Matvezy in #1917

Full Changelog: v0.64.0...v0.64.1

v0.64.0

💪 Added

YOLO26 in inference

Legendary effort of @probicheaux 🏅 and models team made it! Shortly after release, we have a support for new Ultralytics model - YOLO26. If you are interested - learn more about the model, or you can cut the line and jump straight into testing the model on Robolfow Platform.

🔎 How to use the new model?

• Train the model on your data using Robolfow Platform.
• Use it in Workflows or directly in inference

As easy as that 😀

# install inference - uv pip install inference or uv pip install inference-gpu
import cv2
import supervision as sv
from inference import get_model

model = get_model("<id-of-your-yolo26-model>", api_key="<your_api_key>")
image = cv2.imread("<your-image-path>")

predictions = sv.Detections.from_inference(model.infer(image)[0])

annotator = sv.BoxAnnotator()
annotated_image = annotator.annotate(image.copy(), predictions)

New VLMs 🤝 Workflows - Gemini 3 Flash and Claude Opus 4.5

Workflows ecosystem is constantly growing 🤯 Thanks to @yeldarby contributions, we are now able to expose new VLMs in Workflows - plus, for both Gemini and Claude models family, Roboflow now provides managed API keys - so Roboflow account alone is enough - no manual setup of your API keys.

🔧 Bug fixes

• Adds support for selectors in Union[List[...], Selector(...)] patterns in Workflows Execution Engine by @joaomarcoscrs in #1764

🚧 Maintanence

• Fix assertions in GPU tests and add fixture to clean-up memory by @PawelPeczek-Roboflow in #1904
• Fix assertions in GPU tests by @PawelPeczek-Roboflow in #1905
• add yolo and rfdetr search terms to obj-det and keypoint-det workflow blocks by @mkaic in #1906
• Fix documentation links by @sergii-bond in #1908

🥼 New inference - progress update

• Rename inference-exp -> inference-models by @PawelPeczek-Roboflow in #1851
• Fix missing changes of inference-exp to inference-models by @PawelPeczek-Roboflow in #1902
• Fix inference-models again by @PawelPeczek-Roboflow in #1903

🏅 New Contributors

• @sergii-bond made their first contribution in #1908

Full Changelog: v0.63.5...v0.64.0