Issues

• [Tracking] Skill Indexer: retrieval efficiency and accuracy

  area:knowledgekind:tracking
  InferMatrixLab/InferRSI#255 · Filter by author Dong1017 opened on Oct 29, 2026
  5

• [RFC] Online quantization knowledge, skills and Skill Indexer selection

  area:knowledgekind:rfc
  InferMatrixLab/InferRSI#254 · Filter by author zengchuang-hw opened on Oct 29, 2026
  1

• [Tracking] SkillEval evolution: local evidence, PR review and optional LLM judge

  area:knowledgekind:tracking
  InferMatrixLab/InferRSI#253 · Filter by author Dong1017 opened on Oct 29, 2026
  8

• [Tracking] Minimal engineering benchmarks shared by Workbench, SkillEval and SearchSpace experiments

  area:workbenchhelp wantedkind:tracking
  InferMatrixLab/InferRSI#252 · Filter by author Dong1017 opened on Oct 29, 2026
  2

• task: parallel optimization rounds with orthogonal lanes for multi-agent harnesses

  area:processarea:workbenchhelp wantedkind:task
  InferMatrixLab/InferRSI#249 · Filter by author david6666666 opened on Oct 29, 2026
  2

• task: measure whether InferRSI helps Claude Code and Codex optimize inference tasks

  area:knowledgearea:processarea:workbenchhelp wantedkind:task
  InferMatrixLab/InferRSI#248 · Filter by author david6666666 opened on Oct 29, 2026
  3

• task: configure a GitHub Actions release bot

  area:processkind:task
  InferMatrixLab/InferRSI#238 · Filter by author zengchuang-hw opened on Oct 29, 2026

• task: pre-registered campaign on experience and context compaction for agents on the #75 task (Qwen3-TTS 0.6B, RTX 4090), with exploratory results

  area:knowledgekind:rfc
  InferMatrixLab/InferRSI#220 · Filter by author ZX271848 opened on Oct 28, 2026
  1

• [Bug] Admission writer: misleading merge cancellations and repeated HTTP 422 invalidation failures

  area:knowledgekind:task
  InferMatrixLab/InferRSI#203 · Filter by author Dong1017 opened on Oct 28, 2026
  6

  

• [Research Radar] AI inference performance & RSI — daily categorized insights

  area:knowledgekind:investigation
  InferMatrixLab/InferRSI#177 · Filter by author david6666666 opened on Oct 27, 2026
  5

• bug: Drama3 does not fuse NPU normalization operators

  area:workbenchbugkind:investigation
  InferMatrixLab/InferRSI#166 · Filter by author amy-why-3459 opened on Oct 27, 2026
  1

• [RFC] Agent plugins: a consistent InferRSI workflow in Codex, DSH and openJiuwen

  area:workbenchkind:rfc
  InferMatrixLab/InferRSI#152 · Filter by author zengchuang-hw opened on Oct 26, 2026

• [RFC] Hardware-aware optimization goals throughout the RSI workflow

  area:processarea:workbenchkind:rfc
  InferMatrixLab/InferRSI#143 · Filter by author amy-why-3459 opened on Oct 25, 2026
  7

• [RFC] GDN hybrid decider kernels: close HF→vLLM/SGLang gap (causal_conv1d + FlashInfer/Triton GDN)

  area:knowledgekind:rfc
  InferMatrixLab/InferRSI#141 · Filter by author hsliuustc0106 opened on Oct 25, 2026
  1

  

• acceptance: HunyuanImage-3.0 DiT 1024×1024 single-image latency optimization / local L20X

  area:workbenchkind:acceptance
  InferMatrixLab/InferRSI#125 · Filter by author zengchuang-hw opened on Oct 23, 2026
  48

  

• [Status 02] InferRSI evolution: ac702729 → 48e0d589 (2026-10-23)

  area:processkind:status
  InferMatrixLab/InferRSI#116 · Filter by author Dong1017 opened on Oct 23, 2026
  4

• Acceptance: Qwen3-Omni full streaming speech on two L20X GPUs

  area:workbenchkind:acceptance
  InferMatrixLab/InferRSI#112 · Filter by author amy-why-3459 opened on Oct 23, 2026

• [Tracking] SearchSpace distillation coverage and knowledge consolidation

  area:knowledgekind:tracking
  InferMatrixLab/InferRSI#91 · Filter by author Dong1017 opened on Oct 22, 2026
  7

• acceptance: Qwen3-TTS 0.6B C1 RTF optimization on RTX 4090

  area:workbenchkind:acceptance
  InferMatrixLab/InferRSI#75 · Filter by author godnight opened on Oct 21, 2026
  1

• acceptance: Qwen3.8 FP8 vs vLLM 0.32.0 — ≥25% TTFT or TPOT gain, ≤5% regression

  area:workbenchkind:acceptance
  InferMatrixLab/InferRSI#67 · Filter by author jiangkuaixue123 opened on Oct 21, 2026
  330

• [Roadmap] InferRSI: reusable knowledge, verifiable inference optimization

  area:processkind:roadmap
  InferMatrixLab/InferRSI#61 · Filter by author Dong1017 opened on Oct 21, 2026
  3

• acceptance: Qwen3.8-27B single-H20 BF16 text correctness — phase 1

  area:workbenchkind:acceptance
  InferMatrixLab/InferRSI#54 · Filter by author jiangkuaixue123 opened on Oct 21, 2026

• task: evolve a Qwen3.8-27B H20 engine with InferRSI; phase 1 FP8 text service

  area:workbenchkind:task
  InferMatrixLab/InferRSI#53 · Filter by author jiangkuaixue123 opened on Oct 21, 2026
  1

• [Status 01] InferRSI evolution: initialization → ac702729 (2026-10-20)

  area:processkind:status
  InferMatrixLab/InferRSI#41 · Filter by author Dong1017 opened on Oct 20, 2026
  1

• task: add MLX backend discovery and seed source-grounded MLX knowledge for Apple Silicon

  area:knowledgekind:task
  InferMatrixLab/InferRSI#38 · Filter by author shihongzhi opened on Oct 19, 2026