Stay Up-to-Date with Spheron's Latest Innovations

🚀 SPON Token is Now Supported Everywhere

We're excited to announce that SPON is now accepted across the entire Spheron ecosystem!

Whether you're using the Console, Supernoderz, CLI, or Protocol SDK — you can now deposit SPON and use it to pay for any deployment on the network.

If you're a Provider or a Fizz Node operator, you'll also notice a new update in your dashboard - you can now view and withdraw your earnings in SPON.

For developers: to start using SPON payments in your deployments, make sure you're running:

• Protocol SDK v2.5.0 or higher

• CLI v1.9.0 or higher

🖥️ Fizz Node Improvements

We've made a big push to improve reliability, security, and performance across all Fizz Nodes.

Better monitoring and security

Fizz nodes are now constantly monitored for uptime and performance. If a node repeatedly fails security checks or shuts down leased workloads, it will be blacklisted from the network.

New restrictions to keep the network healthy:

• Minimum hardware

  • Nodes now need at least 8 CPU cores, 16 GB RAM, and 100 GB of storage.

• Minimum internet speed

  • Download speeds must be 50 Mbps or higher to join the network.

Other updates

• Pricing for Apple ARM-based devices has been refreshed for CPU, RAM, and storage.

• Upload and download speeds are now tracked separately.

• If the number of “Claimed GPUs” is higher than the actual available GPUs, the claimed count will still be used.

• Correct OS detection for Apple Silicon devices (no more being labeled as “Linux” incorrectly).

• If Docker Hub is down, fizz nodes will fallback to pull images from quay.io instead.

🔧 Updates for Providers

Better monitoring and communication

• Kubernetes providers now get faster, more accurate resource metrics.

• Provider events now use a scalable publish/subscribe system for better reliability.

• Deployment data is fetched from the subgraph first for speed, with RPC calls as a fallback.

🖥 Console App Upgrades

We've made the Console even smoother to use:

• New deployment progress tracker

  • See each step of your deployment as it happens, and get a clear error message if something fails.

• Fiat on-ramp support

  • You can now deposit USD via credit card using CopperX, alongside crypto payments.

• Faster marketplace

  • The marketplace now loads 70% faster than before.

• Better billing view

  • See your total available balance, locked funds, and used amounts all in one place.

⚡ Slark Node Enhancements

• Rewards for Fizz Nodes are now calculated and distributed much faster and more accurately.

• You can now see the percentage of points you've been slashed (if any) over the past 5 days, right on the My Fizz page. This should help you track your performance and take action if needed.

📦 Protocol SDK & Core Updates

Introducing ICL YAML v2

We've introduced ICL YAML v2 - a cleaner, more compact format for defining deployments.

This update removes extra nesting and merges resources, pricing, and placement into a single service definition.

Key Changes

• profiles and placement removed

  • all configs now live under each service.

• expose → port_policy

  • simplified port exposure.

• Resources inline

  • CPU, memory, storage, and GPU requirements now directly in services.

• GPU vendor/model support

  • filter deployments to specific GPU models.

• Pricing inline

  • now inside the service block.

• Simpler deployment block

  • duration, mode, and attributes only.

Example Migration

v1

version: "1.0"
services:
  ollama-test:
    image: ghcr.io/open-webui/open-webui:ollama
    pull_policy: IfNotPresent
    expose:
      - port: 8080
        as: 8080
        to:
          - global: true
profiles:
  name: ollama-testing
  duration: 1h
  mode: fizz
  compute:
    ollama-test:
      resources:
        cpu:
          units: 2
        memory:
          size: 4Gi
        storage:
          - size: 50Gi
placement:
  westcoast:
    attributes:
      desired_fizz: "0xfizznode"
    pricing:
      ollama-test:
        token: uSPON
        amount: 0.5
deployment:
  ollama-test:
    westcoast:
      profile: ollama-test
      count: 1

v2

version: "2.0"
services:
  ollama-test:
    image: ghcr.io/open-webui/open-webui:ollama
    pull_policy: IfNotPresent
    port_policy:
      - port: 8080
        as: 8080
    replica: 1
    resources:
      cpu:
        units: 2
      memory:
        size: 4Gi
      storage:
        size: 50Gi
      gpu:
        units: 1
        attributes:
          vendor:
            nvidia:
              - model: rtx3060
    price:
      token: uSPON
      amount: 1
deployment:
  duration: 1h
  mode: fizz
  attributes:
    desired_fizz: "0xfizznode"

Check out the docs for full details on the new v2 ICL yaml here

Other Updates

• Updated Coinbase Smart Wallet integration to v1.1, improving security, reliability, and wallet compatibility.

• Improved inventory fetching for Fizz and Providers on the marketplace

• Added support for the req_vram attribute, so deployments align better with GPU capabilities.

• Added Thirdweb SDK support in the Fizz Node app, so you can connect more wallet types beyond MetaMask..