-
Notifications
You must be signed in to change notification settings - Fork 289
Pull requests: SemiAnalysisAI/InferenceX
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
[Klaud Cold] Enact the September 8, 2026 DeepSeek-V4-Pro Single-turn 8k1k deprecation / 执行 2026 年 9 月 8 日 DeepSeek-V4-Pro 单轮 8k1k 场景下线
#2921
opened Sep 9, 2026 by
functionstackx
Collaborator
Loading…
fix(evals): preserve nested lm-eval artifacts for sweep summaries
#2920
opened Sep 9, 2026 by
Ankur-singh
Collaborator
Loading…
5 of 9 tasks
perf(sglang): use DSpark6 for B300 DSV4 AgentX / B300 DSV4 AgentX 使用 DSpark6
full-sweep-enabled
#2918
opened Sep 9, 2026 by
nvpohanh
Collaborator
Loading…
[AMD][MI35X] Bump glm5.2-fp4-mi355x-sglang-agentic-mtp image to v0.5.19-rocm10-20260908
full-sweep-fail-fast
#2917
opened Sep 9, 2026 by
jiaryang
Collaborator
Loading…
1 of 2 tasks
[H3 CI] add hardware sites and serving matrices / 接入硬件站点与服务测量
#2916
opened Sep 9, 2026 by
edwingao28
Collaborator
•
Draft
6 of 9 tasks
Turned on the v2 fused top-k for GLM-5.x on ROCm
#2915
opened Sep 9, 2026 by
EricKing626
Collaborator
•
Draft
perf(amd): switch DSV4 ATOM AgentX to native RCCL DEP / 将 DSV4 ATOM AgentX 切换到原生 RCCL DEP
agentx
AgentX benchmarks, recipes, and infrastructure
AMD
full-sweep-enabled
#2912
opened Sep 9, 2026 by
yhl-amd
Collaborator
Loading…
7 tasks done
[AMD][Qwen3.5] Update qwen3.5-fp4-mi355x-sglang-agentic-mtp SGLang ROCm image to v0.5.19-rocm720-mi35x-20260908
full-sweep-fail-fast
#2906
opened Sep 9, 2026 by
yichiche
Collaborator
Loading…
1 task
Optimize GLM-5.2-MXFP4 SGLang Agentic Performance on MI355X
agentx
AgentX benchmarks, recipes, and infrastructure
AMD
full-sweep-enabled
#2900
opened Sep 8, 2026 by
jiejingzhangamd
Collaborator
Loading…
fix(runners): include all ten MI355X runners
#2895
opened Sep 8, 2026 by
cquil11
Collaborator
Loading…
[H3 CI]: add video serving metrics and GPU energy [DO NOT MERGE] / 新增视频服务指标与 GPU 能耗（暂勿合并）
#2894
opened Sep 8, 2026 by
edwingao28
Collaborator
•
Draft
6 of 9 tasks
fix(runners): import B300 container images on the login host
#2893
opened Sep 8, 2026 by
cquil11
Collaborator
Loading…
chore(dsv4): rerun GB300 TRT-LLM AgentX / 重新运行 GB300 TRT-LLM AgentX 配置
#2891
opened Sep 8, 2026 by
RohitNagraj
Collaborator
Loading…
3 of 9 tasks
perf(agentx): sweep Kimi K3 with LMCache dev114 / 使用 LMCache dev114 扫描 Kimi K3
agentx
AgentX benchmarks, recipes, and infrastructure
AMD
full-sweep-enabled
#2889
opened Sep 8, 2026 by
hyukjlee
Collaborator
Loading…
[AMD][MI35X] 0907 DSV4 sglang mtp agentic
AMD
full-sweep-enabled
#2885
opened Sep 8, 2026 by
1am9trash
Collaborator
Loading…
[Klaud Cold] glm5.2-fp8-b300-sglang-agentic-mtp: day-zero GLM-5.2 FP8 SGLang AgentX MTP on B300 / B300 上 GLM-5.2 FP8 SGLang AgentX MTP 首发配方
full-sweep-enabled
#2884
opened Sep 8, 2026 by
functionstackx
Collaborator
Loading…
1 task
[Klaud Cold] Update minimaxm3-fp4-b300-vllm-agentic-mtp vLLM image to nightly-8a728663c1c3eeace834a95f5654fa653cc1998c and move to cluster:b300-dsxe
full-sweep-enabled
#2883
opened Sep 8, 2026 by
functionstackx
Collaborator
Loading…
1 task
[Klaud Cold] Update qwen3.5-fp4-b300-sglang-agentic-mtp SGLang image to nightly-dev-cu13-20260907-30705c00 and move to cluster:b300-dsxe
full-sweep-enabled
#2882
opened Sep 8, 2026 by
functionstackx
Collaborator
Loading…
1 task
[Klaud Cold] Update qwen3.5-fp8-b300-sglang-agentic-mtp SGLang image to nightly-dev-cu13-20260907-30705c00 and move to cluster:b300-dsxe
full-sweep-enabled
#2881
opened Sep 8, 2026 by
functionstackx
Collaborator
Loading…
1 task
[Klaud Cold] Update glm5.2-fp4-b300-sglang-agentic-mtp SGLang image to nightly-dev-cu13-20260908-20ca564b and move to cluster:b300-dsxe
full-sweep-enabled
#2880
opened Sep 8, 2026 by
functionstackx
Collaborator
Loading…
1 task
[Klaud Cold] Update qwen3.8next-fp4-b300-sglang-agentic-mtp SGLang image to dev-cu13-qwen38-next-local (2026-09-07, digest-pinned)
full-sweep-enabled
#2879
opened Sep 8, 2026 by
functionstackx
Collaborator
Loading…
1 task
dsv4-fp4-b300-sglang-agentic-hicache-mtp: restore MegaMoE FP4 acts, retarget to cluster:b300-dsxe / 恢复 MegaMoE FP4 激活路径，并迁移至 cluster:b300-dsxe
full-sweep-enabled
#2878
opened Sep 8, 2026 by
yhyang201
Collaborator
Loading…
[Klaud Cold] Update qwen3.8next-fp8-h200-sglang-agentic-mtp SGLang image to dev-cu13-qwen38-next-local (2026-09-07, digest-pinned)
full-sweep-enabled
#2877
opened Sep 8, 2026 by
functionstackx
Collaborator
Loading…
1 task
[Klaud Cold] Update qwen3.8next-fp8-h100-sglang-agentic-mtp SGLang image to dev-cu13-qwen38-next-local (2026-09-07, digest-pinned)
full-sweep-enabled
#2876
opened Sep 8, 2026 by
functionstackx
Collaborator
Loading…
1 task
[Klaud Cold] Update minimaxm3-fp8-h100-vllm-agentic-mtp vLLM image to nightly-d9105ea8001e0a6d77a96327d17515bb5791fb36
full-sweep-enabled
#2874
opened Sep 8, 2026 by
functionstackx
Collaborator
Loading…
1 task
Previous Next
ProTip!
Filter pull requests by the default branch with base:main.