RallyCompute
.ai
Opening room…
Room
Local room
Coordinator
No coordinator
Model
Execution graph
Shard pipeline
Placement
Worker capacity
Prompting
Context is selected when shards are assigned.
Code
Load files to choose model context.
0 files selected
File preview
Select a file name to preview its contents.
View or edit coding prompt templates
Select files to estimate context.
Generated patch
Staged diff
No staged changes.
Idle
-
Result
Latest execution
Generation
-
Tokens-
Total time-
Effective tok/s-
Advanced
Compute-
WebRTC-
Payload-
KV cache-
Effective time/token-
Pipeline-
First token-
Decode-
-
-
-