Tasks and inputs
SAM 3.1 is listed for Image segmentation, Mask generation, Object detection. It accepts text, image, video and produces mask, bounding box. Use the publisher’s input format and preprocessing for your chosen deployment.
SAM 3.1 finds and tracks objects in images and video, returning bounding boxes and segmentation masks through Meta Model API or its checkpoint runtime.
The specification score summarizes recorded features and limits. It is not a benchmark result.
Pricing conditions: Meta Model API charges $2.50 per 1,000 segmented images or $0.20 per 1,000 video frames. Downloadable checkpoints require their own runtime and licence acceptance. Segmentation is not billed per text token.
Recorded modality support shows what this model can accept and produce. An undocumented modality is shown as unknown rather than assumed to be unsupported.
Accepts
Produces
facebook/sam3.1 is a Hugging Face repository returned by the public Mask generation task feed. Its recorded download count is 84,698 and it has 496 likes when the repository snapshot was collected. Popularity is useful for discovery, but it is not a performance ranking or a guarantee of suitability.
This record is categorized for Mask generation. A task category describes the intended machine-learning problem; it does not prove that every version, checkpoint, quantization, or deployment method produces the same quality. Teams should read the model card and test representative inputs before choosing it.
The public repository lists checkpoint as its library metadata. Its recorded tags include checkpoint, sam3_video, sam3.1, mask-generation, en, license:other, region:us. These source details are retained so readers can trace this directory entry back to the repository instead of relying on copied descriptions.
Use inputs that resemble deployment traffic, including difficult and out-of-domain cases. Compare output correctness, the response format, latency and memory requirements with a simple baseline under the same conditions.
The Hugging Face task assignment records image as input and mask as output for facebook/sam3.1. This interface summary comes from the repository’s recorded Mask generation task classification. It helps readers distinguish a text, image, audio, video, document, embedding, or prediction workflow without inferring undocumented API behavior. Model wrappers can expose different preprocessing and response formats, so verify the upstream model card and the exact runtime before deployment.
Explore facebook models and the mask generation collection. Use the comparison tool to inspect a second candidate.
SAM 3.1 is now also available on Meta Model API as sam-3.1. It accepts a short text phrase with an image or video and returns boxes and pixel masks; video requests track objects across frames. The hosted API is narrower than the checkpoint runtime’s full prompting interface. Image segmentation costs $2.50 per 1,000 images, while video costs $0.20 per 1,000 frames. Keep these units separate when budgeting a clip. Downloadable facebook/sam3.1 weights remain a separate deployment option with their own licence and access conditions.
Compare the model on the work you need to complete.
SAM 3.1 is listed for Image segmentation, Mask generation, Object detection. It accepts text, image, video and produces mask, bounding box. Use the publisher’s input format and preprocessing for your chosen deployment.
Measure a completed task using the prompts, documents and tools your application will use. Include reasoning and retries in the result.
The weights are downloadable under the publisher’s model licence. Check runtime support and size the complete model and working context for your expected concurrency.
These records share a directory category. Check the source card, output type and deployment requirements before treating a model as a replacement.
briaai
briaai/RMBG-1.4 is a source-linked Hugging Face repository assigned to Image segmentation. Its recorded task interface accepts image and produces segmentation mask.
briaai
briaai/RMBG-2.0 is a source-linked Hugging Face repository assigned to Image segmentation. Its recorded task interface accepts image and produces segmentation mask.
chendelong
Hugging Face repository for Image segmentation.
CIDAS
CIDAS/clipseg-rd64-refined is a source-linked Hugging Face repository assigned to Image segmentation. Its recorded task interface accepts image and produces segmentation mask.
cocktailpeanut
Hugging Face repository for Image segmentation.
Hugging Face repository for Image segmentation.