Issues

• Cloud: no API for the credit balance (GET /me/usage from the SDK docs returns 404), and the USD price of a credit appears only in examples

  superlinked/sie#928 · Filter by author mrhwang-dev opened on Nov 9, 2026

• Cloud chat/completions: reasoning_effort is rejected for every model we tried, response_format json_object works only on some, and /v1/models doesn't say which

  superlinked/sie#927 · Filter by author mrhwang-dev opened on Nov 9, 2026

• Cloud: GLM-5.3-Flash and gemma-4-31B-it are listed in /v1/models, but generation returns 503 transport_failure (GLM's :relay profile works)

  superlinked/sie#925 · Filter by author mrhwang-dev opened on Nov 9, 2026

• Measured-only placement planner for worker pools (sie_placement)

  superlinked/sie#903 · Filter by author krisztian-gajdar opened on Nov 6, 2026
  4

• feat: support respanai/span-02 typed decision model

  superlinked/sie#892 · Filter by author svonava opened on Nov 5, 2026

• Docling Cloud extraction returns /dollarsign glyph names in text and table cells

  superlinked/sie#868 · Filter by author svonava opened on Oct 30, 2026

• Cloud first-use: GLiNER2.5-Decide waits 21s before PROVISIONING; clarify readiness and latency expectations

  superlinked/sie#851 · Filter by author svonava opened on Oct 25, 2026
  4

• perf(server): keep CUDA graph replay when LoRA adapters are loaded

  superlinked/sie#840 · Filter by author svonava opened on Oct 20, 2026

• design(helm): define separation for operator-supplied dependency credentials

  superlinked/sie#780 · Filter by author krisztian-gajdar opened on Oct 17, 2026

• MUVERA: compute the encoding on the GPU, not on the inference thread's CPU

  superlinked/sie#443 · Filter by author fm1320 opened on Sep 30, 2026

• GLiClass: large separate-label-group requests can wait minutes in the queue under mixed load

  superlinked/sie#390 · Filter by author svonava opened on Sep 28, 2026
  3

• Batcher: a request that reaches an idle worker waits about 4 ms before dispatch

  superlinked/sie#373 · Filter by author svonava opened on Sep 25, 2026

• Model catalogue - larger open weight models and more powerful GPU

  superlinked/sie#230 · Filter by author ndjones opened on Aug 5, 2026
  3

• Your project is on StackMap — a curated map of the AI stack

  superlinked/sie#217 · Filter by author hoghweed opened on Jul 21, 2026
  2

• ask about this project

  superlinked/sie#191 · Filter by author TentenMarchhhh opened on Jun 12, 2026
  1

• Consider using pre-built Flash Attention kernels via kernels

  superlinked/sie#179 · Filter by author sayakpaul opened on May 28, 2026
  2