市值: $3.3646T 0.850%
體積(24小時): $107.4504B -22.260%
  • 市值: $3.3646T 0.850%
  • 體積(24小時): $107.4504B -22.260%
  • 恐懼與貪婪指數:
  • 市值: $3.3646T 0.850%
加密
主題
加密植物
資訊
加密術
影片
頭號新聞
加密
主題
加密植物
資訊
加密術
影片
bitcoin
bitcoin

$106900.362344 USD

0.81%

ethereum
ethereum

$2539.350639 USD

-0.90%

tether
tether

$1.000181 USD

0.00%

xrp
xrp

$2.355654 USD

-1.42%

bnb
bnb

$651.274881 USD

0.03%

solana
solana

$168.837259 USD

-1.00%

usd-coin
usd-coin

$0.999858 USD

-0.01%

dogecoin
dogecoin

$0.228762 USD

0.62%

cardano
cardano

$0.745002 USD

-0.36%

tron
tron

$0.269741 USD

0.78%

sui
sui

$3.835781 USD

-0.99%

chainlink
chainlink

$15.762179 USD

-2.73%

avalanche
avalanche

$22.438476 USD

-0.40%

stellar
stellar

$0.287075 USD

-0.30%

hyperliquid
hyperliquid

$26.277397 USD

-2.29%

加密貨幣新聞文章

CoinMarketCap揭示了使用LLMS回答有關每個令牌的首要問題的新AI工具

2025/05/21 02:29

越來越多的加密公司正在尋找整合AI的新方法。 5月20日,星期二,CoinMarketCap啟動了AI工具,為用戶提供了有關平台上列出的所有代幣的擴展信息。

CoinMarketCap揭示了使用LLMS回答有關每個令牌的首要問題的新AI工具

Crypto firms are continually finding new ways to integrate artificial intelligence (AI) into their products and services. On Tuesday, May 16, CoinMarketCap launched an AI tool that gives users expanded information about all tokens listed on the platform.

加密公司不斷尋找將人工智能(AI)整合到其產品和服務中的新方法。 5月16日,星期二,CoinMarketCap啟動了一個AI工具,為用戶提供了有關平台上列出的所有代幣的擴展信息。

The AI agents, which are powered by a large language model (LLM) and informed by CoinMarketCap data, explain price movements, offer price predictions, track social sentiment, provide news, and deliver general information about a given token.

由大型語言模型(LLM)提供動力並通過CoinMarketCap數據告知的AI代理商,解釋價格變動,提供價格預測,跟踪社交情緒,提供新聞並提供有關給定令牌的一般信息。

CMC AI is now live on all Coin Detail Pages.Get instant answers about price movements, token fundamentals, and sentiment with a single click.

CMC AI現在在所有硬幣詳細信息頁面上進行。就可以單擊一點擊的價格變動,代幣基礎和情感的即時答案。

No more navigating multiple sources to understand what's happening with your assets. pic.twitter.com/78K6q9l2eC

不再需要多個來源來了解您的資產正在發生的事情。 pic.twitter.com/78k6q9l2ec

“In this first phase of CMC AI, we’re focusing on delivering insights where users need them most—directly on token pages,” said David Salamon, Chief Product Officer at CoinMarketCap. “Our AI is purpose-built for crypto, trained on our extensive market data, and designed to surface insights when users need clarity about specific cryptocurrencies.”

CoinMarketCap首席產品官David Salamon說:“在CMC AI的第一階段中,我們專注於提供最需要用戶的見解 - 直接在令牌頁面上。” “我們的AI是針對加密貨幣的專門製造的,對我們廣泛的市場數據進行了培訓,並在用戶需要清楚特定加密貨幣時旨在表現出洞察力。”

The goal is to enable users to get all the information they need on just one site. This helps users find the info they need more easily, without having to gather data from multiple sources.

目標是使用戶僅在一個站點上獲取所需的所有信息。這可以幫助用戶更輕鬆地找到所需的信息,而無需從多個來源收集數據。

How CoinMarketCap’s AI works

CoinMarketCap的AI如何工作

In a press release shared with crypto.news, CoinMarketCap explained how the new model functions. The tool interfaces with a large language model, such as OpenAI’s o3 reasoning model, providing it with a prompt that includes the latest price data.

在與Crypto.news共享的新聞稿中,CoinMarketCap解釋了新模型的功能。該工具與大型語言模型(例如OpenAI的O3推理模型)的互動,並為其提供了包括最新價格數據的提示。

Once the results are generated, all users who click on one of the questions will see the same output. The responses are not generated in real time but are updated periodically. For major tokens, the AI updates answers every 30 minutes. For smaller tokens, updates are triggered if the price moves more than 2% within one hour.

生成結果後,所有單擊一個問題之一的用戶都將看到相同的輸出。響應不是實時生成的,而是定期更新的。對於主要令牌,AI更新每30分鐘答案。對於較小的令牌,如果價格在一小時內提高2%以上,則會觸發更新。

This model enables users to get instant answers without waiting for AI models to generate responses in real time. It also helps reduce API call costs for CoinMarketCap.

該模型使用戶可以立即獲得答案,而無需等待AI模型實時生成響應。它還有助於降低CoinMarketCap的API呼叫成本。

Still, it’s important to note that LLMs do not always provide accurate responses and can be prone to hallucinations.

儘管如此,重要的是要注意,LLMS並不總是提供準確的響應,並且可能容易出現幻覺。

免責聲明:info@kdj.com

所提供的資訊並非交易建議。 kDJ.com對任何基於本文提供的資訊進行的投資不承擔任何責任。加密貨幣波動性較大,建議您充分研究後謹慎投資!

如果您認為本網站使用的內容侵犯了您的版權,請立即聯絡我們(info@kdj.com),我們將及時刪除。

2025年05月21日 其他文章發表於