Sui price

SuiFest 2025 wrapped last week, and Singapore was buzzing with Sui energy. Over 4,000 builders, partners, and community members came together to celebrate Sui’s momentum, from BTCfi to gaming to next-gen wallets. Here’s what went down 👇

Somnia Network Game · Metaverse Infrastructure Perspective @Somnia_Network is an EVM-compatible L1 that boasts ultra-fast processing (theoretical 1.05 million TPS) and sub-second finality, targeting fully on-chain real-time applications, especially large-scale games and the metaverse. From the early launch, it secured significant partnerships, including a $270 million ecosystem fund from Improbable and MSquared, as well as participation from Google Cloud validators, establishing a foundation of "performance and capital." In the first 36 days of actual mainnet operation, there were 840 million transactions and 300,000 unique addresses, along with a pipeline of over 60 projects spanning gaming, DeFi, and AI. However, the current real-time TPS in practical use remains between 12 and 87, accounting for only about 0.01% of total capacity, with low decentralization indicated by 36 validators and a Nakamoto coefficient of 4, as well as limitations such as community participation driven by airdrop incentives. In summary, "the technology is powerful and the story is compelling, but the 'proof' is still in the early stages." The core technology of Somnia consists of two main components. First, the MultiStream consensus separates data production from consensus by allowing each validator to maintain independent data chains, culminating in a structure that confirms on a modified PBFT-PoS chain. This enables maximum bandwidth utilization and reduces contention issues in parallel models, even in scenarios like gaming tournaments or large minting events where transactions converge on the same contract. Second, the custom KV storage called IceDB aims for deterministic access times in the range of 15 to 100 ns through LSM and immutable section design, enhancing "cost predictability" via gas metering based on cache hits/misses. Additionally, the native compilation of EVM bytecode (AOT/JIT mixed) allows for hash and lookup operations to be executed at the hardware level, achieving speeds close to C++ while maintaining Solidity code. The gap between benchmarks and reality is evident. Stress tests on the development network showed 1.05 million TPS for ERC-20 transfers and 300,000 TPS for NFT minting, with block times at 0.1 seconds and average latency around 1 second. On the mainnet, peak TPS reached 134,000, but real-time performance during regular periods remains in the double digits. The average fee is approximately $0.00013, making it extremely low, and minting and contract calls also remain under a cent. This suggests that while there appears to be sufficient "capacity" for large events, the stability in a continuously high-load environment, such as launching games with millions of concurrent users, has yet to be validated in practice. In terms of competitive positioning, Somnia's stance is clear. Compared to Solana, it offers similarly low fees but claims advantages in sequential execution optimization for game logic that tends to converge at a single point, such as large-scale concurrent minting and reward distribution, rather than the parallel model's vulnerability to contention. However, in terms of actual cumulative transactions, validator distribution, and ecosystem depth, it still struggles to compete with Solana. When compared to ImmutableX, the zero gas minting of ZK rollups and Ethereum security are attractive, but Somnia emphasizes on-chain composability (DeFi + game combinations) and ultra-high throughput at the L1 level. In contrast to Sui, its complete EVM compatibility lowers the entry barrier, while Sui's object parallelism may be more advantageous for tasks with lower correlation. In summary, Somnia differentiates itself with EVM friendliness + sequential execution specialization + predictable gas for a "game-oriented on-chain fairness." On-chain metrics still exhibit a strong speculative and early-stage character. The TVL has surged to about $2.46 million, but the absolute scale is small, with DEX accounting for over 85%. While trading on the derivative DEX (Somnex) has increased, confirming trader influx, the liquidity and composability cases for supporting an AAA-grade game economy are just beginning to take shape. From a user perspective, there was an onboarding spike with 300,000 unique addresses and a maximum of 100,000 DAU in the first month, but recent figures have moderated. On the development side, 435,000 contract deployments (5,300 deployers) in the first month indicate vitality, but the exposure of core low-level technologies (MultiStream, IceDB) as open source is limited. Partnerships and capital lay a solid foundation of trust. Google Cloud encompasses validators, data, and security (Mandiant), while the $270 million from Improbable and MSquared supports onboarding game studios through accelerators (Dream Catalyst) and incubators (Dreamathon). Tools like Sequence and thirdweb, along with collaborations with Yuga Labs, strengthen a hybrid strategy that connects web2 distribution channels (Steam, Epic) with web3 asset capabilities. However, reliance on enterprises like Google creates a tension with the narrative of decentralization. Risks can be summarized in three areas. First, lack of decentralization: 36 validators and a Nakamoto coefficient of 4 are vulnerable to censorship and outages, leading to low trust. Second, insufficient validation of real-world use and scalability: while peaks have been shown, there are no precedents for sustaining over 100,000 TPS in practice. Third, token dilution: with an FDV/MC of 6.24 times (16% circulating), there is significant selling pressure due to unlocks. The community has also shown a tendency towards reward-seeking rather than intrinsic participation, leading to potential exit risks after incentives end. Nevertheless, a sufficient runway ($270 million), EVM friendliness, and game-specialized design present clear opportunities. If Somnia can increase decentralization to over 100 validators and a Nakamoto coefficient of over 10 in the next 6 to 12 months, and successfully launch 1 to 2 AAA titles on the mainnet, proving sustained average TPS of over 100,000 and a TVL of over $100 million, it could become a meaningful option to fill the "gap between Solana, Sui, and IMX." Conversely, if delays in game launches, stagnation in decentralization, and selling pressure from unlocks prevent liquidity and activity from increasing, it risks remaining "another L1 with theoretical TPS ahead." In conclusion, Somnia is a promising candidate with clear technical merits. Its sequential execution optimization, IceDB-based deterministic performance, and complete EVM compatibility reflect a distinct design philosophy aimed at tackling the "concurrent surge" in gaming and the metaverse. However, it is premature to declare it a core infrastructure until the three requirements of decentralization, real-world scale, and token economics are met. The current assessment is that it is a "Credible Contender," and the real turning point will be determined by the sustained TPS, validator structure, TVL, DAU, and game launch performance over the next year.

BNB is now the 3rd most staked asset (in market cap).