Segment Anything 2
Free.ai (self-hosted)
·
image_enhance
·
~1000 Koffice FilterName image
Segment Anything 2 is a beeld opskaalerder gebou deur Meta. Sterkste by Object selection, masking, video object tracking.. Self-hosted on Free.ai GPUs — runs free against your daily token pool (1,000 tokens per beeld). Released under Apache 2.0 — commercial use permitted on Free.ai.
Gebruik via API
OpenAI- versoenbaar met REST API. Genereer 'n sleutel en noem hierdie model in sekondes.
curl -X POST https://api.free.ai/v1/image/enhance/ \
-H "Authorization: Bearer sk-free-..." \
-H "Content-Type: application/json" \
-d '{"model":"sam-2","prompt":"your prompt here"}'Soortgelyke modelle
Vrae wat dikwels gevra word
Segment Anything 2 upskaals en slyp lae-resolusiebeelde terwyl dit detail bewaar. Oplaai 'n klein of wasige foto en Segment Anything 2 keer 'n hoër-oplossing-weergawe terug wat geskik is vir druk, groot vertoon of verdere redigering.
2× and 4× are the standard targets; some configurations support 8×. Final size is capped at 8192×8192 pixels on Free.ai to keep memory bounded.
PNG, JPG, WebP, HEIK en BMP tot 10 mb. Animeerde GIF's word verwerkde raam-by-raamwerk.
Ja, moderne verbeteraars sluit gesig-'nware herstel in. Vir baie verdorwe portrette gee die toegewyde / image/ resore-face/ nutsprogram die skoonste resultaat. Segment Anything 2 hanteer algemene tonele en gesigte in een verbygaan.
Segment Anything 2 loop eers op ons eie GPU's teen jou daaglikse vry poel; dan R2 500 betaalde aandenkings.
Segment Anything 2 hallusines geloofwaardige besonderhede by die opskaalse step ← dit is nie'n feilbare hulpmiddel nie. Vir die meeste verbruikersgebruik (web → druk, mobiele → werkskermpapier, argief skandeer → vertoon) is die resultaat oortuigend.
Krisp op groot teks + eenvoudige logos; klein fyn druk kan op aggressiewe (4x+) instellings kraai. Gebruik 2× wanneer teks getrouheid die meeste saak maak.
Ja verblyd / batch/ aanvaar 'n gids. Uitsets land in /accote /?tab=historie met die oorspronklike lêernaam agtervoegsel van ←-uplundized ${. Vir kabinet- boombehoud gebruik die API.
Yes — POST to /v1/image/enhance/ with model="Segment Anything 2" and your image. /api/ has the full reference + auth.
Same policy as the rest of Free.ai — self-hosted models stay on our GPUs; premium pass through with a DPA; uploads expire on the share-window schedule (24h anon, 7d signed-in). We do not train on your images.
Yes. Free.ai grants commercial use of upscaled images. You must own the source you uploaded.
5–20 seconds at 2×, 15–60 seconds at 4× depending on source resolution and GPU load. Premium models tend to finish on the lower end of those ranges.