personaplex · Issues· 70 open
Open on GitHubLocally synced open issues (discussions stay on GitHub)
- #107
Feature request: deterministic verbatim utterance mode for fixed announcements
Updated Aug 22, 2026 - #106
Service-Duplex-Bench
questionUpdated Aug 20, 2026 - #105
--cpu-offload crashes at warmup: KV cache allocated on meta device for offloaded layers ("Cannot copy out of meta tensor")
Updated Aug 12, 2026 - #96
DGX Spark / GB10: PersonaPlex runs but local Web UI audio latency continuously increases
bugUpdated Aug 6, 2026 - #3
PersonaPlex produces choppy/unusable audio on DGX Spark (GB10)
Updated Jul 24, 2026 - #71
MLX support for Apple Silicon
Updated Jun 27, 2026 - #99
No Retention of system prompt post the context window
questionUpdated Jun 20, 2026 - #97
The model cannot distinguish its role
Updated May 22, 2026 - #93
Tool call support?
questionUpdated May 9, 2026 - #95
Feature request: typed user turns mid-session on /api/chat
questionUpdated May 9, 2026 - #94
Smaller Checkpoint
questionUpdated May 8, 2026 - #66
Connecting RAG and tool calls
questionUpdated May 7, 2026 - #67
[Question] Details on negative-duration silence for barge-in/interruption simulation in training data
questionUpdated Apr 23, 2026 - #26
Can we train it on other languages?
questionUpdated Apr 17, 2026 - #85
Speaker option for this model
questionUpdated Apr 14, 2026