All pull requests

Pull requests

• feat(config): add TokenSpeed Qwen3.8-Flash-Next NVFP4 TP1 AgentX / feat(config): 新增 TokenSpeed Qwen3.8-Flash-Next NVFP4 TP1 AgentX 配置

  #4691·tuanzhangCS opened Oct 31, 2026·7 of 11 tasks·

  ·

• Add Muse Spark Open GB200 SGLang TP8 configurations / 新增 Muse Spark Open GB200 SGLang TP8 配置

  non-canary-full-sweep-enabled
  #4689·RohitNagraj opened Oct 31, 2026Collaborator·

  ·
  6

• Let the PR sweep priority classifier return minimaxm31flash and glm5.3 / 让 PR sweep 优先级分类器支持 minimaxm31flash 和 glm5.3

  #4687·zijiexia opened Oct 30, 2026Collaborator·

  ·
  3

• [AgentX] MiniMax-M3.1-Flash GB300 SGLang: update config / [AgentX] MiniMax-M3.1-Flash GB300 SGLang：更新配置

  full-sweep-fail-fast
  #4686·zijiexia opened Oct 30, 2026Collaborator·

  ·
  11

• [Klaud Cold] Update DeepSeek-V4.1-Flash FP4 MI355X vLLM AgentX DSpark image to nightly-335dd77b / [Klaud Cold] 将 DeepSeek-V4.1-Flash FP4 MI355X vLLM AgentX DSpark 镜像更新至 nightly-335dd77b

  full-sweep-fail-fastklaud-finished
  #4685·Klaud-Cold opened Oct 30, 2026Collaborator·

  ·
  11

• [NVIDIA][AgentX] Qwen3.8-Flash-Next B300 SGLang / [NVIDIA][AgentX] Qwen3.8-Flash-Next B300 SGLang

  full-sweep-fail-fast
  #4683·hshrivastava-droid opened Oct 30, 2026Collaborator·

  ·
  7

• Add Kimi-K3 B300 AgentX on the DSXE EFA cluster / 在 DSXE EFA 集群上新增 Kimi-K3 B300 AgentX

  #4681·mispa-ms opened Oct 30, 2026Collaborator·8 of 12 tasks·

  ·
  3

• [Klaud Cold] Update Qwen3.5 FP8 MI355X ATOM AgentX MTP image to nightly_202610291449 / [Klaud Cold] 将 Qwen3.5 FP8 MI355X ATOM AgentX MTP 镜像更新至 nightly_202610291449

  full-sweep-fail-fastklaud-finished
  #4677·Klaud-Cold opened Oct 30, 2026Collaborator·

  ·
  11

• [AgentX] MiniMax-M3.1-Flash GB300 SGLang: add TP2 and TP4 shortest-prefill-first points / [AgentX] MiniMax-M3.1-Flash GB300 SGLang：新增 TP2 和 TP4 的 shortest-prefill-first 点

  full-sweep-fail-fast
  #4676·ispobock opened Oct 30, 2026Collaborator·

  ·
  6

• [AgentX] MiniMax-M3.1-Flash B300 SGLang: add TP2 and TP4 shortest-prefill-first points / [AgentX] MiniMax-M3.1-Flash B300 SGLang：新增 TP2 和 TP4 的 shortest-prefill-first 点

  full-sweep-fail-fast
  #4675·ispobock opened Oct 30, 2026Collaborator·

  ·
  8

• [NVIDIA][AgentX] Retune GB300 DSV4.1-Flash Dynamo + SGLang HiCache disagg points (stacked on #4595) / [NVIDIA][AgentX] 重新调整 GB300 DSV4.1-Flash Dynamo + SGLang HiCache 分离式点位（基于 #4595）

  #4673·weireweire opened Oct 30, 2026Collaborator·

  ·
  1

• Combine #4537 and #4463 for MiniMax-M3.1-Flash MI355X vLLM: K7 at C16/C32, CPU offload at C64 / 合并 #4537 与 #4463：MiniMax-M3.1-Flash MI355X vLLM C16/C32 使用 K7、C64 使用 CPU 卸载

  full-sweep-enabled
  #4671·Fangzhou-Ai opened Oct 30, 2026Collaborator·

  ·
  9

• [AgentX] Retune Muse Spark Open GB200 SGLang per concurrency / [AgentX] 按并发重新调优 Muse Spark Open GB200 SGLang

  full-sweep-fail-fast
  #4670·shyeh25 opened Oct 30, 2026Collaborator·4 of 11 tasks·

  ·
  4

• Add DeepSeek-V4.1-Flash MI355X vLLM TP8 c1 monokernel scan on latest nightly / 新增 DeepSeek-V4.1-Flash MI355X vLLM TP8 c1 monokernel 单点扫描（最新 nightly）

  full-sweep-fail-fast
  #4666·Fangzhou-Ai opened Oct 30, 2026Collaborator·

  ·
  10

• [AgentX] Qwen3.5-397B-A17B B200 vLLM: add NVFP4 TP2 points with MTP and host KV offload / [AgentX] Qwen3.5-397B-A17B B200 vLLM：新增 NVFP4 TP2 测试点（MTP + 主机 KV offload）

  #4665·aniskumar-nv opened Oct 29, 2026·4 of 6 tasks·

  ·
  2

• Select golden AL curves by verifier and add SPEED-Bench collectors / 按验证器选择黄金 AL 曲线并新增 SPEED-Bench 收集脚本

  #4663·Ankur-singh opened Oct 29, 2026Collaborator·

  ·
  7

• MiniMax-M3.1-Flash GB300 vLLM AgentX: update agentic config / MiniMax-M3.1-Flash GB300 vLLM AgentX：更新 agentic 配置

  full-sweep-enabled
  #4662·hshrivastava-droid opened Oct 29, 2026Collaborator·4 of 9 tasks·

  ·
  6

• Standardize golden AL curve names and selection / 规范黄金 AL 曲线命名与选择

  #4661·Ankur-singh opened Oct 29, 2026Collaborator·

  ·
  16

• [VideoGenX] verify GPU ownership and seal execution evidence / [VideoGenX] 校验 GPU 归属并封存执行证据

  #4659·edwingao28 opened Oct 29, 2026Collaborator·3/5·3 of 10 tasks·

  ·

• [VideoGenX] measure requests and validate generated media / [VideoGenX] 测量请求并校验生成媒体

  #4658·edwingao28 opened Oct 29, 2026Collaborator·2/5·3 of 10 tasks·

  ·

• [VideoGenX] define reproducible H3 workload sets / [VideoGenX] 定义可复现的 H3 负载集合

  #4657·edwingao28 opened Oct 29, 2026Collaborator·1/5·4 of 10 tasks·

  ·

• [PowerX] collect GLM-5.2 B200 Dynamo-SGLang AgentX power / [PowerX] 采集 GLM-5.2 B200 Dynamo-SGLang AgentX 功耗

  full-sweep-fail-fast
  #4656·edwingao28 opened Oct 29, 2026Collaborator·7 of 11 tasks·

  ·
  4

• feat(ci): report-only AgentX speculative acceptance gate / feat(ci)：仅作报告的 AgentX 投机解码接受长度门禁

  #4654·adibarra opened Oct 29, 2026Collaborator·7 of 11 tasks·

  ·
  15

• [Docs] Explain when to use append-only / [Docs] 说明何时使用 append-only

  #4653·edwingao28 opened Oct 29, 2026Collaborator·

  ·

• [Docs] Clarify that the draft KV-cache dtype is not restricted / [Docs] 澄清 draft KV-cache dtype 不受限制

  #4640·nvpohanh opened Oct 29, 2026Collaborator·

  ·