Moondream 2
Free.ai (self-hosted)
·
vision
·
~500 Prometium image
Moondream 2 is a gesigsmodel gebou deur M87 Labs. Sterkste by Low-VRAM image captioning, fast OCR fallback, edge VLM deployments.. Self-hosted on Free.ai GPUs — runs free against your daily token pool (500 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/describe/ \
-H "Authorization: Bearer sk-free-..." \
-H "Content-Type: application/json" \
-d '{"model":"moondream2","prompt":"your prompt here"}'Vrae wat dikwels gevra word
Moondream 2 is a gesigsmodel gebou deur M87 Labs. Sterkste by Low-VRAM image captioning, fast OCR fallback, edge VLM deployments.. Self-hosted on Free.ai GPUs — runs free against your daily token pool (500 tokens per beeld). Released under Apache 2.0 — commercial use permitted on
Moondream 2 works well for Low-VRAM image captioning, fast OCR fallback, edge VLM deployments.. Try the sample prompts above to see its style.
Ongeveer 500 Prometings per gemiddelde boodskap. Vry rekeninge kry 'n 30 000-totken daaglikse bad wat bedek self-geëts gesels; premies modelle is betaal-as-je-go, met 'n goeie top-ups van R1,4.
Dit hang af van die taak. /chat/compare/laat jy dieselfde por aan Moondream 2 en enige ander model kant-side ← vergelyking is die vinnigste manier om te besluit.
Yes. Outputs are yours - Free.ai does not claim rights to anything you generate.Die onderliggende model is Apache 2.0-gelak.
32,768 tokens.
Antwoorde stroom form-by-totken binne ~1 sekonde. Totaal reageertyd hang af van lengte en modelgrootte 539 klein modelle stroom vinniger, grens modelle ruil spoed vir diepte.
Ja. Getekende-in gebruikers sien elke geselsie in /accot/?tab=histories. Jy kan asook deel 'n een-skakel kopie van enige gesprek deur die Deel knoppie.
Free.ai does not train models on your conversations. Self-hosted models stay on our GPUs. Premium models route to the upstream provider for inference.
Yes. POST to /v1/chat/ with model="moondream2" and a messages array. Streaming SSE is supported. Full reference: /api/.
Moondream 2 is Apache 2.0-licensedmet 2B parameters. See /models/moondream2/ for setup notes and our open-source repos at github.com/freeaigit.
Vry rekeninge kry 'n daaglikse bad van 30 000 totken. Wanneer dit opraak, begin die bo-ope begin by $1, betaal-as-Jy-go Ã2m geen intekening nodig.