SGLang adds Jev-style decision-making modes to standard large language models, allowing them to make direct selections without fine-tuning.
1 hours ago
Beating AI Insight News Brief: Large model inference and deployment framework SGLang has added native decision-making APIs, enabling off-the-shelf LLMs and multimodal models like Qwen3.8-27B to directly perform selection, judgment, and scoring without modifying weights or undergoing retraining. When developers input the current context and several candidate answers, the model directly returns the probability of each option, eliminating the need to first generate a text segment for subsequent parsing. This functionality leverages the "next-token probability" that large language models inherently compute. For example, if candidate answers are A, B, and C, a standard chat model would continue generating text, while SGLang directly reads the model’s probabilities for these options to determine its preferred choice. The model itself remains unchanged; only the method of reading outputs is adjusted. SGLang has demonstrated this capability with a *Pokémon FireRed* demo using Qwen3.8-27B. The model decides whether to attack, switch Pokémon, or heal based on real-time game state, with each decision taking less than 100ms, and successfully cleared the Elite Four and Champion in one run. Since Qwen3.8-27B natively supports image input, this approach can also handle multimodal decision-making. Additionally, SGLang has added a new /v1/systemone endpoint, compatible with the TypeSafe SDK used by Jev. Applications already using this SDK can switch their service address to their own SGLang instance to continue making calls. The new feature is now available in the nightly build and scheduled for official release alongside version v0.5.21.
Analysis: The threat posed by a stronger U.S. dollar to Bitcoin may be overestimated, as the two assets have a weak long-term correlation.
3 minutes ago
Goldman Sachs: US pensions may sell off $33 billion in stocks by the end of September.
3 minutes ago
Jumper launches early access testing for Perps, initially rolling out to the top 1000 users ranked by XP.
3 minutes ago
US SEC Plans to Update Transfer Agent Rules; Wall Street Must Avoid 'Paperwork Crisis' in Blockchain Era
3 minutes ago
Amid growing calls for regulation of prediction markets, Polymarket has rolled out a series of user protection and anti-addiction measures.
3 minutes ago
Former New York State Governor: If Democrats win control of Congress, crypto regulatory progress could be rolled back.
3 minutes ago
Hot feeds
A trader profits $448K by monitoring #Binance's new listings!
2024.12.13 17:37:29
Last week, funds have flowed into #Bitcoin, #Ethereum, and #Hyperliquid.
2024.12.16 14:48:36
A $PEPE whale that had been dormant for 600 days transferred all 2.1T $PEPE($52M) to a new address.
2024.12.14 10:35:27
When Elon Musk tweeted about Moltbook, the meme coin MOLT experienced a short-term 30% price surge, hitting a new all-time high of $114 million.
2026.01.31 18:37:29
A smart #AI coin trader made $17.6M on $GOAT, $ai16z, $Fartcoin,$arc.
2025.01.05 16:05:18
A sniper earned 2,277 $ETH ($8.3M) trading $SHIRO within 18 hours!
2024.12.03 23:09:08
MoreHot Articles

How did I turn $1,000 into $30,000 with smart money?
2024.12.09

10 promising AI Agent cryptos
2024.12.05

The 30-Year-Old Entrepreneur Behind Virtual, a Multi-Million Dollar AI Agent Society
2025.01.22

10 smart traders specializing in MEMEcoin trading on Solana
2024.12.09

A trader lost $73.9K trading memecoins in just 3 minutes — a lesson for us all!
2024.12.13

What is $SPORE? Let us take you through the on-chain records to show you how it works.
2024.12.25

