-
Notifications
You must be signed in to change notification settings - Fork 269
Pull requests: SemiAnalysisAI/InferenceX
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
feat(agentx): retune Kimi-K3 FP4 MI355X ATOM DSpark recipe on _0821
agentx
AgentX benchmarks, recipes, and infrastructure
AMD
full-sweep-enabled
#2716
opened Aug 24, 2026 by
zejunchen-zejun
Collaborator
Loading…
fix: fold DP_SIZE into single-node num_gpus for per-GPU throughput
#2715
opened Aug 24, 2026 by
aistackdev
Loading…
Refresh the DSv4 MI355X SGLang AgentX arm to v0.5.18 and split TP4/TP8 by concurrency
agentx
AgentX benchmarks, recipes, and infrastructure
AMD
full-sweep-enabled
#2713
opened Aug 24, 2026 by
karverma-amd
Collaborator
Loading…
4 of 5 tasks
config(dsv4): constrain OpenMP threads for GB300 AgentX / 限制 GB300 AgentX 的 OpenMP 线程数
full-sweep-enabled
#2711
opened Aug 24, 2026 by
RohitNagraj
Collaborator
Loading…
4 of 9 tasks
[AgentX] DeepSeek-V4 B300 SGLang update
full-sweep-enabled
#2704
opened Aug 21, 2026 by
yhyang201
Collaborator
Loading…
[AgentX] DeepSeek-V4 B300 SGLang update
full-sweep-enabled
#2701
opened Aug 21, 2026 by
Oasis-Git
Collaborator
Loading…
[AMD][MI35X] Serve Qwen3.5 MXFP4-AttnFP8-V2 on the MI355X SGLang single-node arms
full-sweep-fail-fast
#2695
opened Aug 20, 2026 by
yichiche
Collaborator
Loading…
[AMD][MI35X] Add HiCache and TP2/EP1 arms to the Qwen3.5 MXFP4 MI355X AgentX sweep
full-sweep-enabled
#2693
opened Aug 20, 2026 by
yichiche
Collaborator
Loading…
[AMD] Refresh DeepSeek-R1 MI355X SGLang image / [AMD] 更新 DeepSeek-R1 MI355X SGLang 镜像
all-evals
Expand eval selection to every fixed-sequence config
full-sweep-fail-fast
#2691
opened Aug 20, 2026 by
Oseltamivir
Collaborator
Loading…
[Power] Add optional B200 and B300 multinode telemetry / 添加可选 B200 和 B300 多节点功耗采集
#2688
opened Aug 20, 2026 by
edwingao28
Collaborator
•
Draft
[Power] Bulk-enable GB multinode telemetry / 批量启用 GB 多节点功耗采集
#2687
opened Aug 20, 2026 by
edwingao28
Collaborator
•
Draft
[Power] Require GLM-5.2 and add DSV4 H200 telemetry / 强制 GLM-5.2 并新增 DSV4 H200 功耗采集
#2684
opened Aug 19, 2026 by
edwingao28
Collaborator
•
Draft
[Power] Wire H200 multinode AgentX power / 接入 H200 多节点 AgentX 功耗链路
#2683
opened Aug 19, 2026 by
edwingao28
Collaborator
•
Draft
Test AIPerf empty-content TTFT on MiniMax TRT
full-sweep-enabled
#2682
opened Aug 19, 2026 by
cquil11
Collaborator
Loading…
[DO NOT MERGE][Test] DSV4 FP4 B300 SGLang AgentX DEP8 c512 on nightly (HiCache + MegaMoE FP4-act)
sweep-enabled
#2680
opened Aug 19, 2026 by
yhyang201
Collaborator
Loading…
[AMD] [AGENTX] Kimi-K3 Perf Tuning 2.0
agentx
AgentX benchmarks, recipes, and infrastructure
AMD
#2674
opened Aug 19, 2026 by
ajith-sirra-amd
Collaborator
Loading…
feat(config): add GLM-5.2 NVFP4 B200 Dynamo-SGLang AgentX recipes / 新增 GLM-5.2-NVFP4 B200 Dynamo-SGLang AgentX 配方
sweep-enabled
#2673
opened Aug 19, 2026 by
Ankur-singh
Collaborator
Loading…
5 of 9 tasks
[AMD] [AGENTX] Minimax-M3 Perf Tuning
agentx
AgentX benchmarks, recipes, and infrastructure
AMD
full-sweep-enabled
#2671
opened Aug 19, 2026 by
ajith-sirra-amd
Collaborator
Loading…
Perf tuning for GB300 agentx Kimi K3
full-sweep-enabled
#2669
opened Aug 19, 2026 by
wzhao18
Collaborator
Loading…
feat(glm52-agentx): Add B300 Dynamo+TRT-LLM AgentX recipes
full-sweep-enabled
#2666
opened Aug 19, 2026 by
RohitNagraj
Collaborator
•
Draft
4 of 9 tasks
perf(gb300): Add more AgentX vLLM MTP aggregate and P/D recipes / 增加更多 GB300 AgentX vLLM MTP 聚合与 P/D 配方
agentx
AgentX benchmarks, recipes, and infrastructure
full-sweep-enabled
#2665
opened Aug 19, 2026 by
ivanium
Collaborator
Loading…
[WIP] Add GLM-5.2-NVFP4 B300 AgentX recipes
full-sweep-enabled
NVIDIA
#2654
opened Aug 18, 2026 by
csahithi
Collaborator
Loading…
Add Qwen3.5 FP4 B300 TRT-LLM Disaggregated AgentX / 新增 Qwen3.5 FP4 B300 TRT-LLM 分离式 AgentX
full-sweep-enabled
#2653
opened Aug 18, 2026 by
richardhuo-nv
Collaborator
Loading…
Previous Next
ProTip!
no:milestone will show everything without a milestone.