DeepSeek V4 Flash 0423

DeepSeek V4 Flash is an efficiency-optimized Mixture-of-Experts model from DeepSeek with 284B total parameters and 13B activated parameters, supporting a 1M-token context window. It is designed for fast inference and...

La descrizione sottostante proviene da OpenRouter ed è disponibile solo in inglese.

Rilascio

Finestra di contesto

1M token

Output massimo

944K token

Input

testo

Output

testo

Ragionamento

Sì

Storico dei prezzi

• input $0.14 → $0.091 -35% · output $0.28 → $0.182 -35%
• input $0.047 → $0.14 +196% · output $0.095 → $0.28 +196%
• input $0.045 → $0.047 +5% · output $0.45 → $0.095 -79%
• input $0.047 → $0.045 -5% · output $0.45 → $0.45 0%
• input $0.05 → $0.047 -5% · output $0.45 → $0.45 0%
• input $0.14 → $0.05 -64% · output $0.28 → $0.45 +61%
• input $0.091 → $0.14 +54% · output $0.182 → $0.28 +54%
• input $0.05 → $0.091 +82% · output $0.45 → $0.182 -60%
• input $0.091 → $0.05 -45% · output $0.182 → $0.45 +147%

Provider su OpenRouter

17 endpoint servono questo modello. L'uptime indica la percentuale di richieste riuscite su OpenRouter; prezzi per 1M token.