issues Search Results · language:Dune language:JavaScript language:JavaScript language:JavaScript language:Python is:public
Filter by
40.8M results
!-- sentinel:error-fingerprint:1658b78b7836b083 --
Problem
Sentinel captured a deploy_failure error from deploy-status:ahead.
Observed Error
Live tree has 1 commit(s) not on origin/main.; Live tree ...
autonomy
codex
needs-approval
priority:medium
reliability
self-heal:escalate
Thanks for releasing this model. The prosody sounds pretty good overall.
I’m trying to use it for Hindi, but the pronunciation is bit off in some cases.
Are you planning to share more details about the ...
!-- sentinel:error-fingerprint:8be960b955b876de --
Problem
Sentinel captured a health_degradation error from health.
Observed Error
{ engine_version : error: engine_process_stale }
Suspected Area ...
autonomy
codex
needs-approval
priority:high
reliability
self-heal:escalate
Please check the Github page for a better reading experience and more papers.
Time Series
| Title | Date | Comment |
| --- | --- | --- |
| LLM-Powered Multi-Agent System for Automated Crypto Portfolio ...
documentation
Parent: #9
Scope
Add a frontend-admin app validation baseline for app payload consistency without duplicating completed canvas, component
registry, property editor, binding, permission, or mobile preview ...
area:frontend-admin
priority:p1
type:task
Problem
take v3.0.0 ends at behavior-contract delivery and hands off to the runa runtime to sequence the downstream stations
(plan → implement → verify → submit). Runa autonomous sequencing is not yet ...
enhancement
Inside the footer, when hover onto the Cookie Policy nothing being displayed
Please check the Github page for a better reading experience and more papers.
source localization
| Title | Date | Comment |
| --- | --- | --- |
| A Primer on Evolutionary Optimization Frameworks for ...
documentation
Coverage gap in `backend/src/agents/tools.py`:
``` src/agents/tools.py 331 15 95% 56, 334, 394-395, 434-435, 523, 570-571, 674-675, 738, 783, 792-793 ```
Line 56 is the `return [` opening of the `get_tools()` ...
我在双 RTX 2080 Ti 的 SM75 机器上测试时发现,VLLM_SM75_SPEC_SYNC_MODE=nosync 是目前最大的 decode 加速项。
它可以把官方 pure-filler profile 的 decode 速度从大约 50 tok/s 提升到 100.5 tok/s。
但是同一套模型和同一类 chat 请求,在开启 nosync 后会出现真实聊天输出不稳定的问题。比如简单中文状态询问会开始重复标点: ...

Learn how you can use GitHub Issues to plan and track your work.
Save views for sprints, backlogs, teams, or releases. Rank, sort, and filter issues to suit the occasion. The possibilities are endless.Learn more about GitHub IssuesProTip! Restrict your search to the title by using the in:title qualifier.