Livepeer AI (SPE)
Livepeer AI (SPE)
  • Changelog
  • Feature requests
  • Planned
  • Backlog
  • Completed
  • In Progress
  • To do

Feature requests

Discover our plans and suggest new improvements.
Planned

Implement C2PA spec into pipelines

Backlog

Implement AI job verification

Communicate used model parameters back to Gateway

Add custom container support

Add communication protocol for long-running requests

Network Analytics

Provide Gateways with on connection runner GPU stats

Nov 29
Batch Jobs Support

Ensure container cleanup on AI worker shutdown

Improve Developer Experience

Create Generic inference job type

Completed
Batch Jobs Support

Add support for fine-tuned model hot swapping

Sep 2
Improve Developer Experience

Generate Livepeer AI client SDKs

Sep 5
Increase Supply

Allow multiple gpus per capability

Aug 19
Backlog

Communicate AI-Runner errors to the Orchestrator

In Progress

Add Frame Interpolation in go-livepeer

Ensure json string format works in `aiModels`

To do

Black images returned from ai-runner

Create an AI orchestrator control dashboard