Served on MixpeekFaceVerified 2026-11-05
InsightFace buffalo_sc (SCRFD-500MF + MobileFaceNet WebFace600K)
insightface/buffalo_sc
InsightFace buffalo_sc (SCRFD-500MF + MobileFaceNet WebFace600K) is a face recognition model. Served on Mixpeek, scaling to zero when idle.
A Ray Serve app in Mixpeek's inference registry hosts it. It starts at zero replicas and wakes on first use.
New opens the Designer in Studio. For an existing namespace, open it in Studio and use its Designer tab.
No demo for InsightFace buffalo_sc (SCRFD-500MF + MobileFaceNet WebFace600K) yet
A video, image, slideshow or playground goes here once we record one on our own data.
What it is good and bad at
- Best at
- The fastest face detector option (SCRFD-500MF) for high-volume frames where small faces matter less.
- Worst at
- Recall on small faces and identity accuracy: model_zoo lists buffalo_s/sc well below buffalo_l on IJB-C and MR-ALL. Commercial use is not licensed.
- Task
- Face
- Input
- Image, Video
- Output
- Boxes, Keypoints, Vector
- Use cases
- Face search
Templates that use it
No template uses this model yet. Browse templates.
Input and output
Input
- image
- det_size
- 640x640
Output
- boxes
- face bounding boxes with detection scores
- keypoints
- 5 landmarks per face
- vector
- dims
- 512
- distance
- cosine
- type
- dense
How Mixpeek wires it
- Extractors
face_identity_extractor@v1detection_model=scrfd_500m (buffalo_sc detector; embeddings still from buffalo_l)
- Feature URIs
mixpeek://face_identity_extractor@v1/insightface__arcface
- Vector indexes
face_identity_extractor_v1_embedding- Retriever stages
- feature_search
- Combines with
- buffalo_s (variant) is loaded only for the Serve app's embedding_model=arcface_r50, a MobileFaceNet whose vectors are not comparable with buffalo_l's
How it runs
- Status
- Served on Mixpeek Loaded by the insightface__arcface Serve app when a request names detection_model=scrfd_500m or embedding_model=arcface_r50.
- Serve apps
insightface__arcface- Runtime
- insightface.app.FaceAnalysis(name="buffalo_sc") on onnxruntime (buffalo_s for arcface_r50)
- Weights
- ONNX, no remote code
- Parameters
- Not recorded
- Size on disk
- 16 MB
- Preprocessing
- SCRFD detection at 640x640.
- Postprocessing
- Boxes and 5-point landmarks passed to alignment and the recognition model.
- Serving
- both, batch 8
- Hardware
- Runs on CPU (slower without a GPU)
Scale to zero
- Minimum replicas
- 0
- Maximum replicas
- 4
- Idle timeout
- 600 s
- Memory
- 2 GB
- CPUs
- 0.5
- GPUs
- Set at deploy time: hardware_config.gpu_fraction if hardware_config.gpu_concurrency > 0 else 0
- Cold start, node warm
- Not recorded
- Cold start, new node
- Not recorded
Measured cost
Unmeasured
No committed measurement.
Our evaluations
Unmeasured
We have not evaluated this model on our own data yet. Publisher results are in the sources below.
Questions
- What is InsightFace buffalo_sc (SCRFD-500MF + MobileFaceNet WebFace600K) best at?
- The fastest face detector option (SCRFD-500MF) for high-volume frames where small faces matter less.
- What are the limits of InsightFace buffalo_sc (SCRFD-500MF + MobileFaceNet WebFace600K)?
- Recall on small faces and identity accuracy: model_zoo lists buffalo_s/sc well below buffalo_l on IJB-C and MR-ALL. Commercial use is not licensed.
- Can InsightFace buffalo_sc (SCRFD-500MF + MobileFaceNet WebFace600K) be used commercially?
- No. Same terms as every InsightFace pretrained pack: non-commercial research only per the insightface README and model_zoo.
- How does InsightFace buffalo_sc (SCRFD-500MF + MobileFaceNet WebFace600K) run on Mixpeek?
- Loaded by the insightface__arcface Serve app when a request names detection_model=scrfd_500m or embedding_model=arcface_r50.