什么是 Bittensor Subnet 85 Vibe Miner？

Bittensor Subnet 85 Vibe Miner 技能是一个全面的自动化软件包，旨在简化 VidAIo 网络上视频处理矿工的设置和管理。它简化了为视频放大和压缩配置高性能 GPU 实例的复杂过程，确保开发人员和矿工能够以最小的人干预最大化其 TAO 收益。通过利用 Openclaw Skills，用户可以部署生产级的挖掘节点，其中包括关键的性能补丁和优化的工作流程。

该技能专为处理 Subnet 85 的技术细节而设计，例如 VMAF 质量评分和硬件加速编码。无论您是使用 Vast.ai 进行云端挖掘还是使用自己的裸机硬件，该技能都能提供必要的脚本和配置，以保持竞争排名并避免常见的陷阱，如评分衰减或验证器超时。将其集成到您的 Openclaw Skills 库中，即可实现具有现场经过验证的优化的专业级挖掘操作。

下载入口:https://github.com/openclaw/skills/tree/main/skills/maxquick/bittensor-sn85-vibe-miner

安装与下载

1. ClawHub CLI

从源直接安装技能的最快方式。

2. 手动安装

将技能文件夹复制到以下位置之一

优先级：工作区 > 本地 > 内置

3. 提示词安装

将此提示词复制到 OpenClaw 即可自动安装。

Bittensor Subnet 85 Vibe Miner 应用场景

• 自动部署使用 RealESRGAN 算法的视频放大矿工。
• 为 AV1 和 HEVC 格式配置高效视频压缩矿工。
• 集成去中心化存储解决方案（如 Hippius）或 S3 兼容的 BackBlaze B2，用于处理矿工数据。
• 实施关键性能补丁，以在 Bittensor 网络上保持高激励分数。
• 使用 Vast.ai 或本地硬件跨多个 GPU 实例扩展挖掘操作。

1. 代理执行先决条件检查，以验证 GPU 可用性、TAO 钱包余额和存储凭据。
2. 它自动执行环境设置，在 Ubuntu 22.04 上安装 NVIDIA 驱动程序、CUDA、Docker 和必要的仓库。
3. 为 BackBlaze B2 或 Hippius 应用存储配置，以处理视频数据的输入和输出需求。
4. 将关键性能补丁（如 VMAF 计算修复和 video2x 速度优化）应用于源代码。
5. 代理协助创建钱包热钥和子网注册，以便在 Subnet 85 上获取 UID。
6. 通过 PM2 启动挖掘服务，并主动监控 axon 连接和任务处理。
7. 持续的健康检查和激励跟踪确保矿工保持竞争力，并避免在免疫期内被取消注册。

Bittensor Subnet 85 Vibe Miner 配置指南

# 克隆仓库并初始化挖掘环境
git clone https://github.com/vidaio/vidaio-subnet.git
cd vidaio-subnet

# 运行 Openclaw Skills 提供的自动化安装脚本
./scripts/install.sh

# 配置您的存储提供商 (BackBlaze 或 Hippius)
./scripts/storage_setup.sh

# 在 Subnet 85 上注册您的矿工
./scripts/register.sh --netuid 85 --wallet.name  --wallet.hotkey 

Bittensor Subnet 85 Vibe Miner 数据架构与分类体系

Bittensor Subnet 85 "Vibe Miner" Skill

Author: Mark Jeffrey
Description: Complete automated setup for mining Bittensor Subnet 85 (VidAIo) — video upscaling and compression mining with GPU acceleration.

What This Skill Does

Guides users through setting up profitable video processing miners on Bittensor Subnet 85:

• Upscaler miner — Upscale SD→HD, HD→4K using RealESRGAN (video2x)
• Compressor miner — High-quality video compression with AV1/HEVC
• Storage setup — BackBlaze B2 or Hippius decentralized storage
• Monitoring — Rank tracking, incentive monitoring, health checks
• Optimizations — Includes critical patches for VMAF quality scoring and video2x speed

Prerequisites

Required:

1. NVIDIA GPU — RTX 3090/4090 recommended (needs NVENC hardware encoding)
2. TAO Wallet — 1-2 τ minimum (≈$200-400 at current prices)
3. Storage — BackBlaze B2 account OR Hippius access
4. Linux — Ubuntu 22.04+ recommended
5. Skills — Basic command line, SSH access if using Vast.ai

Recommended Platform:

• Vast.ai — Rent RTX 4090 for $0.30/hr ($216/month)
• Alternative — Own hardware or other GPU cloud providers

Expected Returns

Based on live mining data (March 2026):

• Upscaler earnings: ~$14/day (top 5% miners)
• Compressor earnings: ~$9/day (top 65% miners)
• Combined: $23/day ($690/month)
• Break-even: ~2-3 weeks on Vast.ai rental

Returns vary with TAO price, network competition, and subnet emissions.

How to Use This Skill

Agent will guide you through:

1. Creating/importing TAO wallet
2. Setting up GPU instance (Vast.ai or bare metal)
3. Configuring storage (BackBlaze or Hippius)
4. Installing dependencies and mining software
5. Registering on subnet (costs ~0.13 τ per miner)
6. Starting miners and monitoring performance

What Makes This Different

Includes hard-won optimizations:

• ? VMAF quality scoring — Critical patch that many miners miss (causes rank decline)
• ? video2x speed optimization — 10-50x faster than default Python implementation
• ? Vast.ai port workarounds — Handles occupied ports (8080, 11111, 18384)
• ? Score decay awareness — 0.8 decay factor, strategies to maintain rank
• ? Dual miner setup — Both upscaler + compressor for maximum earnings

Agent Instructions

When user requests SN85 mining setup:

Phase 1: Prerequisites Check

1. Verify GPU availability (query user about Vast.ai vs own hardware)
2. Check TAO wallet exists (btcli wallet list)
3. Verify wallet balance (need 1-2 τ minimum)
4. Confirm storage choice (BackBlaze B2 or Hippius)

Phase 2: Environment Setup

1. If Vast.ai: Guide through instance creation (RTX 4090, Ubuntu 22.04, Docker)
2. If bare metal: Verify NVIDIA drivers, CUDA, docker/PM2
3. Clone vidaio-subnet repository
4. Install dependencies (use scripts/install.sh)

Phase 3: Storage Configuration

1. If BackBlaze: Guide through B2 bucket creation, save credentials
2. If Hippius: Set up using hippius skill, configure S3-compatible endpoint
3. Update miner configs with storage credentials

Phase 4: Apply Patches

1. Apply VMAF calculation patch (patches/validator_merger_vmaf.patch)
2. Verify video2x installation (should use CLI, not Python)
3. Test GPU encoding (NVENC support)

Phase 5: Registration

1. Create hotkeys if needed (btcli wallet new_hotkey)
2. Register upscaler on SN85 (costs 0.13 τ)
3. Register compressor on SN85 (costs 0.13 τ)
4. Configure port mappings (Vast.ai: map internal→external)

Phase 6: Start Miners

1. Launch PM2 services (upscaler, compressor, workers)
2. Verify axon connectivity
3. Wait for first tasks (upscaler should receive within 30-60min)

Phase 7: Monitor

1. Check incentive/rank with scripts/monitor.sh
2. Verify VMAF calculation in logs (look for "?? Calculating final video VMAF")
3. Alert if crash loops detected
4. Track earnings over immunity period (7200 blocks ≈ 24h)

Key Files

• SKILL.md — This file
• scripts/install.sh — Full dependency installation
• scripts/register.sh — Subnet registration helper
• scripts/monitor.sh — Status checking and alerts
• scripts/storage_setup.sh — BackBlaze/Hippius configuration
• patches/validator_merger_vmaf.patch — Critical quality scoring fix
• configs/miner_upscaler.json — Upscaler configuration template
• configs/miner_compressor.json — Compressor configuration template

Troubleshooting

Compressor crash loops:

• Check for Python syntax errors in validator_merger.py
• Verify VMAF patch applied correctly
• Look for port conflicts

Low incentive scores:

• Verify VMAF calculation is running (check logs for "?? Calculating")
• Ensure GPU encoding is working (NVENC)
• Check network connectivity (axon reachable)

Deregistration risk:

• Monitor incentive (if below 0.002, at risk)
• Upscaler has priority (better task frequency)
• Consider running upscaler only if capital limited

Safety Notes

• Test on testnet first if unsure
• Start with one miner to verify setup before registering second
• Monitor closely during immunity period (first 24h)
• Backups — Keep wallet mnemonics secure and offline
• Costs — Registration is non-refundable, calculate ROI before committing

Support

This skill packages real mining experience from March 2026. All optimizations are battle-tested.

Common issues solved:

• VMAF calculation disabled by default (causes rank decline)
• video2x Python implementation too slow (causes validator timeouts)
• Vast.ai port conflicts (8080, 11111, 18384 occupied)
• Score decay mechanics (0.8 factor, need frequent validator pings)

Ready to mine? Agent will walk you through each step. Have your TAO wallet and GPU ready.