SmolLM 3 3B (3B) iku a chat model dibangun déning HuggingFace. It accepts up to 32K tokens of context per request. Strongest at Low-memory devices, fast inference, on-device chat.. Self-hosted on Free.ai GPUs — runs free against your daily token pool (100 tokens Per pesen). Relea

SmolLM 3 3B gawé apik kanggo Low-memory devices, fast inference, on-device chat.. Coba conto pitakon ing ndhuwur kanggo ndeleng gayané.

Kira-kira 100 tokens saben pesen rata-rata. Akaun gratis nampa pool 30,000-token saben dina kang nglindhungi chatting kang dihost-kanthi-diri; model premium bayar-kaya-kanggo-mu, karo token top-up saka $1.

It depends on the task. /chat/compare/ lets you send the same prompt to SmolLM 3 3B and any other model side-by-side - comparison is the fastest way to decide.

Ya. Output iku saka sampeyan - Free.ai ora nganggep hak kanggo apa wae sing sampeyan hasilake. The underlying model is Apache 2.0-licensed.

32,768 tokens.

Jawaban stream token-by-token ing ~1 detik. Total wektu jawab gumantung ing dawa lan ukuran model - model cilik stream luwih cepet, model frontier trade kecepatan kanggo jero.

Ya. Pengguna kang wis mlebu bisa ndeleng saben chatting ing /account/?tab=history. Sampeyan uga bisa nyawisaké salinan siji-pandhu arah saka saben chatting liwat tombol Share.

Free.ai ora nglatih model ing konvèkasi sampeyan. Model kang dihost-kanthi-diri tetep ing GPU kita. Model Premium rute menyang penyedia upstream kanggo inferensi.

Ya. POST menyang /v1/chat/ kanthi model "smollm3-3b" lan array pesen. Streaming SSE didhukung. Referensi lengkap: /api/.

SmolLM 3 3B is Apache 2.0-licensedkaro parameter 3B. See /models/smollm3-3b/ for setup notes and our open-source repos at github.com/freeaigit.

Akun gratis bisa nampa 30.000 token saben dina. Nalika iku wis rampung, token top-ups wiwit ing $1, pay-as-you-go - ora ana subscription dibutuhaké.