The 2.78-trillion-parameter Kimi K3 can run on just 8GB of memory, with its developers open-sourcing a lightweight C-based inference engine.
2 hours ago
A developer recently open-sourced the kimi-k3-in-c project, attempting to run the 2.78-trillion-parameter Kimi K3 model on devices with only 8GB of RAM. The project is just 176KB in size, written purely in C99, and does not depend on GPU, CUDA, PyTorch, or BLAS—its model inference can be completed entirely via CPU. This solution leverages the Mixture of Experts (MoE) architecture features of Kimi K3. While the model’s total parameter scale reaches 2.78T, only 16 of the 896 experts per layer are activated. Instead of loading the full ~1.56TB model weights into memory, the developer stores most expert weights on NVMe drives and reads them in real time as needed for inference; some dense trunk layers also adopt layer-by-layer streaming loading. However, the solution still has obvious performance limitations: in 8GB RAM mode, the model takes approximately 32.7 seconds to generate one token, and requires nearly 1.7TB of high-speed storage support. The developer stated that this solution is currently more of an experimental exploration into optimization directions for large model inference infrastructure, and does not have practical production use value. Nevertheless, its approach of "NVMe streaming loading + MoE sparse activation" provides a new idea for the future low-cost operation of ultra-large models.
To avoid a government shutdown, the U.S. Senate passed a temporary funding bill.
7 minutes ago
Attacker of Aztec’s private Rollup bridge transfers another 300 ETH to Tornado Cash, bringing total mixed ETH to 500.
7 minutes ago
TUT surges over 55% in 24 hours, with its market capitalization climbing to $40 million.
7 minutes ago
IMF: Domestic stablecoins may boost demand for U.S. dollar stablecoins
7 minutes ago
Sources: Senior US military officials are seeking ways to disengage from the Iran conflict.
7 minutes ago
JPMorgan Chase: U.S. and European companies' earnings beat expectations, posting strong results in the Q2 earnings season.
7 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

