Morph MORPH
SIGN IN SIGN UP
dhruvhead / InferenceX UNCLAIMED
0 0 96 Python

COMMITS

main
September 7, 2026
J
[Klaud Cold] [AMD] Bump DSV4 MI355X vLLM 8k/1k to the 2026-09-01 nightly / [Klaud Cold] [AMD] 将 DSV4 MI355X vLLM 8k/1k 更新至 2026-09-01 nightly (#2792)
jiacao-amd committed
24a78a3
F
[Klaud Cold] Update dsv4-fp4-mi355x-vllm-agentic-mtp vLLM ROCm image to nightly-e962733e08d10f7ca65dac4df99e116460b8b174 (#2841)
functionstackx committed
9acf24c
September 5, 2026
A
fix(klaud): enforce shippable images and reusable sweeps (#2842)
Alec Ibarra committed
e52162e
A
fix(klaud): release capacity-deferred candidates (#2839)
Alec Ibarra committed
80677d8
A
fix: respect public API caching in Klaud Cold / 修复 Klaud Cold 公开 API 缓存处理 (#2837)
Alec Ibarra committed
fdcca1e
A
fix(klaud): reuse published baselines and monitor owned runs (#2836)
Alec Ibarra committed
4c46551
A
fix(klaud): randomize selection and recheck target capacity (#2834)
Alec Ibarra committed
7bde875
A
fix(klaud): recover overlap reviews and correct agent naming (#2832)
Alec Ibarra committed
54a35ed
K
[AMD][DSV4] Add a DP-attention arm and refresh the MI355X SGLang AgentX key (#2800)
karverma-amd committed
650ee9e
September 4, 2026
A
feat(ci): add autonomous Klaude Cold auto-sweep (#2830)
Alec Ibarra committed
9bf93d0
F
[AMD][MI355X] Bump MiniMax-M3 AgentX to latest ROCm nightly / [AMD][MI355X] 将 MiniMax-M3 AgentX 更新至最新 ROCm nightly (#2825)
Fangzhou Ai committed
d45ff43
W
[CI] Import the DCGM exporter via enroot registry syntax / 用 enroot registry 语法导入 DCGM exporter (#2735)
Wenyao Gao committed
08ccbc6
A
feat(runners): add the B300 DSXE cluster and retire the B300 NV launcher (#2826)
Ankur Singh committed
ee8f457
W
Update Kimi K3 GB300 Agentx (#2811)
Wei Zhao committed
cf08a54
S
perf(agentx): refresh K3 MI355X vLLM recipe with DCP8 MTP arm (#2810)
seungrokj committed
c355f54
September 3, 2026
X
[Klaud Cold] dsv4-fp4-b200-vllm: refresh nightly and use CuTeDSL EP backend / 更新 DSV4 B200 vLLM nightly 并使用 CuTeDSL 专家并行后端 (#2534)
Xin Li committed
d13df37
R
config(dsv4): constrain OpenMP threads for GB300 AgentX / 限制 GB300 AgentX 的 OpenMP 线程数 (#2711)
Rohit Nagraj committed
62c140a
R
feat(config): add GLM-5.2 GB300 AgentX concurrency-1 disaggregated point / 添加 GLM-5.2 GB300 AgentX 并发度 1 分离式配置 (#2720)
Rohit Nagraj committed
f1e2e19
C
fix: align AgentX offload metadata with runtime recipes (#2817)
Cameron Quilici committed
131d052
C
chore(runners): remove decommissioned B300 fleet (#2818)
Cameron Quilici committed
c1746a9
C
fix(config): route all GB300 workloads to batch 1 (#2813)
Cameron Quilici committed
6829ed5
September 2, 2026
A
perf(agentx): refresh GLM-5.2 B200 SGLang curve / 性能：刷新 GLM-5.2 B200 SGLang 曲线 (#2808)
Ankur Singh committed
f2be63a
September 1, 2026
C
Refresh DeepSeek V4 GB300 TRT-LLM AgentX metrics (#2774)
Cameron Quilici committed
5ecf6fd
C
Refresh GLM-5.2 GB300 TRT-LLM AgentX metrics (#2773)
Cameron Quilici committed
e807a52
C
Refresh MiniMax M3 B300 TRT-LLM AgentX metrics (#2772)
Cameron Quilici committed
5de5561
C
Refresh MiniMax M3 B200 TRT-LLM AgentX metrics (#2771)
Cameron Quilici committed
e81d7e0
C
Refresh Qwen3.5 GB300 TRT-LLM AgentX metrics (#2770)
Cameron Quilici committed
0a5d6d3
A
perf(qwen3.5-fp4-b200-sglang-mtp): enable FlashInfer GDN for TP2/EP2 / 为 TP2/EP2 启用 FlashInfer GDN (#2790)
Ankur Singh committed
db7d21f
C
test: remove tautological tests and document test quality [skip-sweep] (#2799)
Cameron Quilici committed
bdca939
R
Add DSV4 B200 disaggregated Dynamo SGLang STP configuration (#2560)
Rohit Nagraj committed
2bd4738