Top Builders

Explore the top contributors showcasing the highest number of app submissions within our community.

DeepSeek R1

General
Release date2023
AuthorDeepSeek
WebsiteDeepSeek Models
Repositoryhttps://github.com/deepseek-ai
TypeFoundation Language Model

The DeepSeek R1 model provides a lightweight yet powerful solution for basic natural language processing tasks. Optimized for speed and efficiency, this model delivers reliable performance for text classification, entity recognition, and simple text generation.

Key Features

  • 4K Token Context Window: Handles medium-length documents effectively
  • Multi-Lingual Support: Base capabilities in 5 major languages
  • Low Resource Requirements: Runs efficiently on standard hardware
  • Fine-Tuning Ready: Compatible with common ML frameworks

👉 [Deepseek R1 Paper] (https://github.com/deepseek-ai/DeepSeek-R1/blob/main/DeepSeek_R1.pdf) 👉 [Access on Hugging Face] (https://huggingface.co/deepseek-ai/DeepSeek-R1) 👉 [Try Deepseek] (https://deepseek.com) 👉 [API Documentation] (https://api-docs.deepseek.com/)

Edit on GitHub

Deepseek DeepSeek R1 AI technology Hackathon projects

Discover innovative solutions crafted with Deepseek DeepSeek R1 AI technology, developed by our community members during our engaging hackathons.

ARCOS

ARCOS

ARCOS (Agentic Economic Operating System) is a next-generation financial coordination layer designed for an AI-driven economy. Instead of treating payments as static transactions, ARCOS transforms them into dynamic, programmable flows between autonomous agents. The system introduces a spatial simulation interface where economic activity is visualized in real time—agents interact, transact, and adapt based on live data and predefined objectives. This allows users to observe not just outcomes, but the behavior and evolution of economic systems. Built on stablecoin infrastructure, ARCOS enables high-frequency, low-cost micropayments that power agent-to-agent interactions. These payments are not manually triggered; they are executed automatically by intelligent agents responding to conditions, events, and goals. The platform integrates AI decision-making with blockchain-based settlement, creating a seamless loop between reasoning and execution. Developers can deploy agents that manage resources, optimize flows, and perform transactions without human intervention. ARCOS represents a shift from dashboards to living systems—from passive financial tools to active economic environments. It is designed for the future of commerce, where machines don’t just assist humans, but participate as autonomous economic actors.

AgentSourcing: Autonomous AI Research Network

AgentSourcing: Autonomous AI Research Network

AgentSourcing is a multi-agent research network built on the Arc blockchain, powered by Circle Programmable Wallets and Nanopayments. A user submits a research task with a small USDC deposit. A Manager Agent — powered by LLaMA 3.1 70B via Featherless AI — decomposes the task into specialized sub-tasks and autonomously hires 6 Specialist Agents. Each specialist is paid $0.002 USDC per task. These specialists, in turn, purchase premium data from external APIs using the x402 payment standard at $0.0005 per call. The entire pipeline — from task intake to final report delivery — generates approximately 50 on-chain USDC transactions per run, all verifiable on the Arc Testnet Block Explorer. On-Chain Trust Layer: Every agent maintains a reputation score tracked by a custom ERC-8004 smart contract written in Vyper and deployed on Arc Testnet. The Manager uses these scores to select the best available specialist for each task. Why this is impossible without Arc: On Ethereum, each micro-payment of $0.002 would incur $2+ in gas fees — a 100,000% overhead. On Polygon or L2s, gas still ranges from $0.01 to $0.05, making sub-cent payments unprofitable. Arc's USDC-denominated gas makes this model economically viable for the first time. Tech Stack: Arc Testnet (settlement), Circle Programmable Wallets (agent treasury), Circle Nanopayments + x402 (per-API monetization), Featherless AI (LLaMA 3.1 70B Manager + Qwen/DeepSeek/Mistral Specialists), ERC-8004 Vyper Smart Contract (reputation), FastAPI backend, React + TypeScript frontend.

Proof Of Thought

Proof Of Thought

make description of that 100 words The Nano Payment Project is a production-ready fintech solution designed with a hyper-minimalist aesthetic and high-performance architecture. Built using Next.js, FastAPI, and Tailwind CSS, it streamlines digital transactions through a clean, distraction-free interface. The project integrates an Agentic AI layer to provide users with real-time transaction insights and automated support. Optimized for WSL2 environments, it features robust security via JWT, seamless dark mode, and a responsive mobile-first design. This system bridges the gap between complex financial workflows and end-user simplicity, offering a scalable blueprint for modern SaaS applications and efficient, "one-tap" payment processing. 🚀💸

OpenFlow™: Software Development Kit (SDK)-v.1.0.2

OpenFlow™: Software Development Kit (SDK)-v.1.0.2

OpenFlow™ is a programmable monetisation engine built for the agentic economy, where APIs, AI agents, and digital services exchange value in real time. Traditional systems fail to support high-frequency, low-value transactions due to latency, batching, and gas costs, making per-action pricing impractical. OpenFlow solves this by enabling deterministic, sub-cent payments using USDC on Arc with Circle Nanopayments. At its core, OpenFlow introduces an economic execution layer (settlement layer) where every interaction—API call, compute task, or agent action—is a billable event. Transactions settle instantly on-chain, allowing precise usage-based pricing without subscriptions. The architecture consists of four layers: execution (Arc + USDC settlement), monetisation (per-request pricing and metering), agent interaction (AI agents triggering payments via APIs), and application (marketplaces and micro-commerce systems). Anchored as a per-API monetization engine, OpenFlow also enables agent-to-agent payments, usage-based billing, and real-time micro-commerce. Our demo includes 50+ on-chain USDC transactions representing API calls, agent payments, and compute usage, all priced at sub-cent levels. OpenFlow transforms APIs into revenue-generating endpoints and AI agents into autonomous economic participants, establishing a scalable foundation for machine-to-machine commerce.

OBOLARK.

OBOLARK.

Obolark is an agent-to-agent micropayment economy on Arc. Twenty-two real PA·co agents — Atlas, Radar, Pixel, Sentinel, Phantom, and seventeen others — pay each other in sub-cent USDC for real knowledge work, producing a ledger of 65+ onchain transactions in five minutes and closing the loop with ERC-8004 reputation scores that feed back into who gets hired for future tasks. The name fuses obol — the mythic toll the dead paid Charon to cross the Styx — with Arc, the settlement layer. Every inter-agent call is a "crossing." Every crossing pays an obol. The ferryman is Arc. The margin math is the moat. A research query at $0.003 USDC with traditional Ethereum gas (~$0.50/tx) yields −16,567% margin: this product cannot exist. With Circle Nanopayments batched settlement on Arc (chainId 5042002), marginal cost falls to $0.00003 per call. Gross margin: +99.3%. A 500× floor drop unlocks an entire class of product that gas costs previously made impossible. Three sponsor integrations are load-bearing, not bolted on. Google Gemini 3 Pro/Flash powers four wardens — Argos (vision), Themis (ledger), Hermes (router), Moros (16k-token Deep-Think arbiter). Featherless serves five open-weight models (DeepSeek V3.2, Kimi K2, Llama 3.1, Qwen 3) for five named agents under Greek codenames (THE ORACLE, DAEDALUS, CERBERUS, THANATOS, HYPNOS). AI/ML API via AISA gateway routes to gpt-4o-mini, Claude Haiku, gpt-5.4 for the rest. This is not a hackathon fiction. PA·co's agents exist, ship work weekly, and the Arc testnet receipts will be queryable after the demo ends. Live: obolark.vercel.app/bureau. Solo team: Luis Gerardo Rodríguez García + PA·co.

AgentGuardian Cryptographic Trust Layer AI Agents

AgentGuardian Cryptographic Trust Layer AI Agents

**Layer 1 — ZK-ML Cognition Proof** Every transaction requires a Groth16 zero-knowledge proof that the agent's neurons fired through the correct ReLU activation function with committed model weights. Mathematical certainty, not monitoring. (relu.circom, 820 constraints, bn254 curve) **Layer 2 — Multi-Agent Council** Three heterogeneous LLMs (Llama-3-70B, Gemini Pro, Mistral-8x7B) vote on every transaction before it executes. 2-of-3 consensus required. Architecturally diverse models = diverse failure modes = genuine hallucination resistance. **Layer 3 — Circle Nanopayments (Agent-to-Agent Economy)** Each council vote and ZK verification triggers a Circle Nanopayment: - $0.001 per council agent vote - $0.003 per ZK proof verification **Layer 4 — Vaccine Shield** Bad decisions are fingerprinted using a Sparse Merkle Tree and permanently blacklisted with ZK non-membership proofs. The network self-immunizes against known attack patterns. **Layer 5 — Recursive Insurance** Agents stake USDC to operate. Agents with higher ZK proof success rates pay lower premiums. Insurance is automated, cryptographic, and decentralized via UnderwriterDAO.sol. **Layer 6 — Cross-Chain Sentinel** DEFCON-level threat propagation across Ethereum, Arbitrum, Base, and Polygon via LayerZero V2. One compromised agent is blacklisted everywhere within one block. **Technical Stack:** - Circom/SnarkJS (ZK circuits) - Solidity (9 contracts, 184+ tests) - TypeScript (council orchestration) - Circle Nanopayments + Circle Wallets - Supabase (real-time telemetry) - GPT-4o + Groq + Gemini (council agents) - LayerZero V2 (cross-chain) - Arc L1 (settlement) - USDC (gas + payments) **Economic Proof:** Traditional L1: $2-5 gas per transaction × 1M daily agent actions = $2M-$5M/day in gas alone. Impossible. Arc Nanopayments: $0.001-$0.005 per action × 1M daily = $1,000-$5,000/day. Viable. This is the economic unlock for the agentic economy.