Ramp Labs Introduces Multi-Agent Memory Sharing Solution, Token Consumption Reduced by Up to 65%

By: theblockbeats.news|2026/04/11 06:19:36
0
Share
copy

BlockBeats News, April 11th, AI infrastructure company Ramp Labs released research results on "Latent Briefing", achieving efficient memory sharing among multi-agent systems through direct compression of large-scale model KV cache, significantly reducing Token consumption without sacrificing accuracy.


In mainstream multi-agent architectures, the Orchestrator decomposes tasks and repeatedly calls Worker models. As the inference chain extends, Token usage exponentially inflates. The core idea of Latent Briefing is to leverage the attention mechanism to identify the truly critical parts in the context, directly discard redundant information at the representation layer, rather than relying on the slow-speed LLM summary or the unstable RAG retrieval.


In the LongBench v2 benchmark test, this method performed remarkably: Worker model Token consumption decreased by 65%, the median Token savings for medium-length documents (32k to 100k) reached 49%, the overall accuracy improved by approximately 3 percentage points compared to the baseline, and the additional time for each compression was only about 1.7 seconds, achieving a speedup of about 20 times compared to the original algorithm.


The experiment used Claude Sonnet 4 as the Orchestrator, and Qwen3-14B as the Worker model, covering various document scenarios such as academic papers, legal documents, novels, and government reports. The research also found that the optimal compression threshold varies depending on task difficulty and document length—difficult tasks are suitable for aggressive compression to filter out speculative reasoning noise, while long documents are more suitable for mild compression to retain scattered key information.

-- Price

--

You may also like

Former ByteDance employee's account: How I started with two Pinduoduo hard drives and made six times the profit with Seagate to achieve financial freedom?

A programmer from a big tech company bought hard drives on Pinduoduo and, following clues, managed to accurately capture the sixfold rising stock Seagate using the "finding daily anomalies + 13F institutional verification" framework, making a wild profit of $400,000 and achieving financial freedom.

MiCA reshuffle begins, Binance temporarily bids farewell to the EU

What Binance leaves behind is not scattered retail investors, but a whole batch of high-value users who are forced to liquidate and have almost nowhere to go.

How does Gate redo "buying and selling stocks" from the cryptocurrency world to the stock market?

The competition logic of exchanges has changed.

Visa and Mastercard join 140 giants to launch a new stablecoin, but the impact on the market landscape may still be limited

As an important milestone event in the stablecoin landscape, OUSD is likely to change the existing stablecoin landscape and significantly increase the adoption rate of stablecoins in the global financial system.

Circle CEO responds to OUSD's challenge: Stablecoins are a winner-takes-all business, and we will not slow down

OUSD was jointly launched by more than 140 giants, causing Circle's stock price to plummet in a single day. Circle's CEO personally wrote a response, clarifying USDC's moat from three aspects: network effects, liquidity, and regulation, and dismantling OUSD's three selling points of "free redemption...

Argentina vs Cape Verde: When a Record-Breaking Legend Meets an Unbreakable Underdog

WEEX exclusive pre-match analysis of Argentina vs Cape Verde, exploring Messi-led Argentina’s dominance and Cape Verde’s historic defensive breakout, with a breakdown of volatility, structure, and match dynamics.

Contents

Popular coins

Latest Crypto News

Read more
iconiconiconiconiconiconicon
Customer Support:@weikecs
Business Cooperation:@weikecs
Quant Trading & MM:bd@weex.com
VIP Program:support@weex.com