Hy4 preview

tencent/hy4-preview

text input · text output · konteks 1,048,576 · output maks 64,000

alatskemapenalarancachetanpa pelatihan
Biaya efektif per 1rb permintaan

biaya = p_in·(input − cached) + p_cache_read·cached + p_out·output + p_req; token tercache adalah prefiks yang sudah disimpan endpoint sticky (part2b §9.3), sehingga harga daftar termurah tidak selalu menghasilkan permintaan termurah.

PenyediaWilayah · kuantisasiKonteks / output maksInput / MOutput / MBaca / tulis cachePer 1rb permintaanUptime 30 hrTTFT p50 / p95Throughput24 jam terakhirFidelitasDataKemampuanStatus
ai.mlglobal · unknown1,048,576 / 64,000$0.876$2.626$0.044 / -$2.233---tidak ada lalu lintas- tanpa pelatihan 30 hrbawaan modelaktif

Mulai cepat

curl https://api.ai.ml/v1/chat/completions \
  -H "Authorization: Bearer $AIML_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"tencent/hy4-preview","messages":[{"role":"user","content":"Say hello"}]}'
Dapatkan kunci di konsol

Log perubahan

  • new_model model tencent/hy4-preview added (active); endpoint ep_aiml_hy4-preview_global (tencent/hy4-preview) added as active; price input 0.876 output 2.626 USD/M 5/10/2026