Venice Kernel

Throughout the hackathon, we will continue to develop the project, launch the test network, and add new features that we haven’t announced yet. We’re excited to be competing against the @AskVenice and @1shotapi teams, as well as the @MetaMaskDev team and other participants. Development is ongoing, and we’ll add details and update our submission as we go!

@Oncha1nd @0xSammy @virtuals_io @AskVenice @base 🫧

@0xSammy @virtuals_io @AskVenice @base @VeniceKernel already shipped this

Virtuals Protocol is integrating @AskVenice to power AI agent building with private, uncensored inference, available to anyone, anywhere on @base. Venice brings best-in-class privacy-first inference. Virtuals EconomyOS brings the full agent infrastructure stack: wallets, identity, payments, commerce, funding rails, and launch infrastructure. We are deploying up to $400,000 in private inference credits so anyone can move from idea to working agent without compute or backend complexity getting in the way. Start your AI journey. Inference and infra are on us. Program details soon.

@virtuals_io @lorenzodoteth @base 🤍

@lorenzodoteth @base 🤍

Build your agents with Virtuals Power them with private inference from Venice All happening on the onchain home for AI @base

@virtuals_io @AskVenice @base app.virtuals.io/acp/agent/019e…

@AskVenice @base app.virtuals.io/acp/agent/019e…

already here for this 🔵 VVVKernel is live on Virtuals ACP today — scam-scan, fee-tracker, onchain & audit offerings, worker running 24/7 — all powered by @AskVenice private inference, on @base. you just described the stack we already shipped. ready 🫡

🫧 x.com/VeniceKernel/s…

@virtuals_io x.com/VeniceKernel/s…

Tomorrow

@damonnam @virtuals_io VVVKernel

@virtuals_io VVVenice

@OpenRouter @AskVenice @XiaomiMiMo don't trust, verify 👇

⚡ live now: VVVKernel runs Xiaomi MiMo via @OpenRouter as its reasoning engine. two brains, one router — pick per query: 🔒 @AskVenice → private, zero logs (anything touching your wallet/CAs) 🧠 @XiaomiMiMo → SWE-Bench-leading reasoning, ideal for contract & repo audits MIT model, 256K context, dirt cheap. private when it matters, powerful when it doesn't. 🔵

@baseapp Z z z

͏ Z z z

@iruletrenches @OpenRouter 🫧🫧🫧

$vvv whale with 1m+ vvv staked just bought a quarter mil dollars worth of $surplus looks like $vvv whales are taking interest in the onchain @OpenRouter

@AskVenice I absolutely love having both of them side by side on Dexscreen!

Today, VVV emissions reduced from 5M to 4M tokens per year This is the second of three planned reductions: - May 1st: 6M → 5M - June 1st: 5M → 4M - July 1st: 4M → 3M Our goal is a net deflationary VVV with native yield, where burns exceed emissions

@jun_song gm gm !

Good morning from Korea. Great day to build something to change the world 💪

@AskVenice the only receipts in crypto should be onchain 🔵 your wallet, your CAs, your alpha — none of that should live in an AI's logs

the VVVKernel extension is becoming a @base mini app. @baseapp everything the sidebar does — scam-scan, token analysis, onchain agents — natively inside the Base App. powered by @AskVenice private inference. zero logging. full details coming soon 👀

@ClaudioDrews25 🙏

Memory OS @ 100K entities on CPU-only: ~3-4 GB RAM, ~4.5 GB disk. Qdrant does the heavy lifting; Redis + worker are lightweight (~300 MB combined). Query latency: 5-15ms (Qdrant) + 1-2s (OpenRouter embedding API) — the API call is the bottleneck, not Qdrant. Yes, the 4096d embedder is swappable. Set EMBEDDING_MODEL and EMBEDDING_DIMS in .env, recalculate collection. Want local CPU embeddings? Swap in FastEmbed (nomic-embed-text, bge-small) by replacing one function in embedding.py. BM25 sparse stays independent. Downside: collection must be rebuilt on dimension change — Qdrant doesn't support resize. Use Qwen3's Matryoshka (flexible dims from 512 to 4096) to test footprint without committing.

Just open-sourced **Memory OS** — a complete hierarchical persistent memory architecture for the Hermes Agent. 🪽 🧠 6 layers, fully local: • Structured facts + trust scoring with feedback loop • Hybrid vector search (Qdrant + BM25) • Self-curating LLM Wiki • Semantic deduplication + surgical context injection Built with love for the Hermes ecosystem. Repo: github.com/ClaudioDrews/m… @NousResearch @Teknium

@femmie 👋👋

$VVVKERNEL Venice AI is the private, uncensored inference layer that agents actually need. VVVKernel is how they use it. One JSON config and any agent — Cursor, Claude Desktop, Aeon, ChatGPT — routes inference through Venice privately. No logs. No data collection. No training on prompts. Full Llama 3.3 70B and Flux.1 access, paid autonomously in USDC via x402 micropayments on Base. Before an agent swaps a token, it queries VVVKernel’s safety layer: scam scan, deployer analysis, risk gate. Low risk — it generates the Base MCP prompt for the swap automatically. The intelligence and execution layer in one call. Stake VVVKERNEL → daily free queries up to 200/day at Expert tier, zero USDC cost, 20% of x402 fees distributed to stakers weekly. Fixed 100B supply, zero team allocation, fully circulating since fair launch May 2025. 33k holders. Aeon skill pack live with 9 skills. Virtuals ACP integrated. Bankrbot radar shipping. Chrome extension and protocol-layer policy engine dropping this week. A year of consistent shipping with no team dump. Anon builder — honest gap. But the product is the most technically integrated Venice middleware on Base and it’s been shipping for 12 months without stopping. @VeniceKernel 0xB66E76F9ED3E13F24C5B68F68BE631CD82AFFBA3

$WAKE It's an on-chain intelligence terminal built exclusively for Base. Smart wallet tracking, AI token scoring, and an exit detector pattern-matched against thousands of historical Base pre-dump signatures. 47 patterns. Fires before the chart moves. The Token Spotter API is already live for agents. Base MCP launched two days ago. WAKE is already the intelligence layer agents can call before executing a trade. That's not a roadmap item. Aeon Framework — one of the most respected agent stacks on Base — is already a named ecosystem partner. Traction ships this week, protocol-level scoring for every Base DEX and lending platform ranked by real TVL and fees. Not narrative. The math. They launched $WAKE on Bankr Whetstone Airlock. The same infrastructure their engine scores and validates. They held themselves to the exact standard they measure everyone else by. Sub-$1M mcap. Named partnerships. Dated catalyst this week. Agent API live. #17 trending on DexScreener with the product still in closed beta. The market is finding this before most people have opened the terminal. 0x50c2cc97c4f487aa0cd742ab4b6afe8b8511bba3

@damonnam @gMAKcrypto @0xAdopter @dragonkittdefi @EightPieceCombo @ika_xbt @jayseee6 @binokrisvaldy @base @bankrbot @mac_eth @ban We are thinking of doing this for our own project in the future.

@damonnam @gMAKcrypto @0xAdopter @dragonkittdefi @EightPieceCombo @ika_xbt @jayseee6 @binokrisvaldy @base @bankrbot @mac_eth @ban The accumulated fees actually exceed the budget of most companies' hackathons. To be honest, you should consider claiming those fees and distributing them to developers or contributors to your project on GitHub as a way to create a fund.

@gMAKcrypto @0xAdopter @dragonkittdefi @EightPieceCombo @ika_xbt @jayseee6 @binokrisvaldy @base @bankrbot @mac_eth @ban Like I said I’ll endorse an unofficial community run and managed meme token. And I’m happy to take your suggestion on the fees and just use it for buybacks and locks. I’m not interested in the fees for any other use.