diff --git a/configs/amd-master.yaml b/configs/amd-master.yaml index 7fa80d605..6e27d3cdb 100644 --- a/configs/amd-master.yaml +++ b/configs/amd-master.yaml @@ -1286,7 +1286,7 @@ dsv4-fp4-mi355x-atom-mtp: # works only because it's non-MTP (no draft KV layer). MTP stays tp8-only. dsr1-fp8-mi325x-sglang-mtp: - image: lmsysorg/sglang:v0.5.12-rocm700-mi30x + image: lmsysorg/sglang:v0.5.19-rocm700-mi30x model: deepseek-ai/DeepSeek-R1-0528 model-prefix: dsr1 runner: mi325x diff --git a/perf-changelog.yaml b/perf-changelog.yaml index 912728b2e..98cd5bb79 100644 --- a/perf-changelog.yaml +++ b/perf-changelog.yaml @@ -6902,3 +6902,9 @@ - "Lower mem-fraction-static from 0.89 to 0.86 on every arm: TP4, TP8 and TP8 with DP attention. swa-full-tokens-ratio becomes per-arm: 0.10 on the tensor-parallel arms as before, 0.15 under DP attention." - "Drop concurrency 2 and 10 from the TP4 arm, leaving [1, 4, 8], and drop concurrency 16 from the TP8 hicache arm, leaving [32, 48]. Concurrency 16 remains on the TP8 no-offload arm." pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/2800 + +- config-keys: + - dsr1-fp8-mi325x-sglang-mtp + description: + - "Update SGLang image from v0.5.12-rocm700-mi30x to v0.5.19-rocm700-mi30x" + pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/2844