💥 Gate Square Event: #PostToWinPORTALS# 💥
Post original content on Gate Square related to PORTALS, the Alpha Trading Competition, the Airdrop Campaign, or Launchpool, and get a chance to share 1,300 PORTALS rewards!
📅 Event Period: Sept 18, 2025, 18:00 – Sept 25, 2025, 24:00 (UTC+8)
📌 Related Campaigns:
Alpha Trading Competition: Join for a chance to win rewards
👉 https://www.gate.com/announcements/article/47181
Airdrop Campaign: Claim your PORTALS airdrop
👉 https://www.gate.com/announcements/article/47168
Launchpool: Stake GT to earn PORTALS
👉 https://www.gate.com/announcements/articl
The DeepSeek online model has been upgraded, the current version number is DeepSeek-V3.1-Terminus.
DeepSeek announced this evening that the online model has been upgraded, with the current version number DeepSeek-V3.1-Terminus, which includes two versions: the thinking model and the non-thinking model, both with a context length of 128k, and it is now available for users to experience online. Among them, the output length of the non-thinking model is set to 4K by default, with a maximum of 8K, while the output length of the thinking model is set to 32K by default, with a maximum of 64K. In terms of pricing, the model charges 0.5 yuan for one million tokens input (cache hit), 4 yuan for cache misses, and 12 yuan for one million tokens output.