Skip to content
Draft
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
15 changes: 10 additions & 5 deletions .github/configs/amd-master.yaml
Original file line number Diff line number Diff line change
@@ -1,5 +1,5 @@
dsr1-fp4-mi355x-sglang:
image: lmsysorg/sglang:v0.5.12-rocm700-mi35x
image: lmsysorg/sglang:v0.5.13-rocm700-mi35x
model: amd/DeepSeek-R1-0528-MXFP4-Preview
model-prefix: dsr1
runner: mi355x
Expand All @@ -11,13 +11,18 @@ dsr1-fp4-mi355x-sglang:
- isl: 1024
osl: 1024
search-space:
- { tp: 4, conc-start: 4, conc-end: 64 }
- { tp: 8, conc-start: 4, conc-end: 64 }
- { tp: 4, conc-start: 4, conc-end: 256 }
- { tp: 8, conc-start: 4, conc-end: 256 }
- { tp: 4, ep: 4, conc-start: 4, conc-end: 256 }
- { tp: 8, ep: 8, conc-start: 4, conc-end: 256 }
Comment thread
cursor[bot] marked this conversation as resolved.
- isl: 8192
osl: 1024
search-space:
- { tp: 4, conc-start: 4, conc-end: 64 }
- { tp: 8, conc-start: 4, conc-end: 64 }
- { tp: 4, conc-start: 4, conc-end: 256 }
- { tp: 8, conc-start: 4, conc-end: 256 }
- { tp: 4, ep: 4, conc-start: 4, conc-end: 256 }
- { tp: 8, ep: 8, conc-start: 4, conc-end: 256 }

# Agentic-coding sweep commented out for this image-bump PR — the
# 10-conc agentic matrix amplifies sweep cost and the bump validation
# only needs the fixed-seq-len throughput shape. Re-enable once the
Expand Down
8 changes: 7 additions & 1 deletion benchmarks/single_node/fixed_seq_len/dsr1_fp4_mi355x.sh
Original file line number Diff line number Diff line change
Expand Up @@ -5,6 +5,7 @@ source "$(dirname "$0")/../../benchmark_lib.sh"
check_env_vars \
MODEL \
TP \
EP_SIZE \
CONC \
ISL \
OSL \
Expand Down Expand Up @@ -37,16 +38,21 @@ fi
# Start GPU monitoring (power, temperature, clocks every second)
start_gpu_monitor

if [ "${EP_SIZE:-1}" -gt 1 ]; then
PARALLEL_ARGS+=( --ep-size "$EP_SIZE" )
fi

set -x
python3 -m sglang.launch_server --model-path=$MODEL --trust-remote-code \
--host=0.0.0.0 --port=$PORT \
"${PARALLEL_ARGS[@]}" \
--tensor-parallel-size=$TP \
--chunked-prefill-size=$PREFILL_SIZE \
--mem-fraction-static=0.8 \
--disable-radix-cache \
--num-continuous-decode-steps=4 \
--max-prefill-tokens=$PREFILL_SIZE \
--cuda-graph-max-bs=128 \
--cuda-graph-max-bs=512 \
--attention-backend aiter \
--kv-cache-dtype fp8_e4m3 $EVAL_CONTEXT_ARGS > $SERVER_LOG 2>&1 &

Expand Down
5 changes: 5 additions & 0 deletions perf-changelog.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -3919,3 +3919,8 @@
- "Also update all configs for DSR1 TRTLLM FP8 to reflect latest released image usage"
pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/1767

- config-keys:
- dsr1-fp4-mi355x-sglang
description:
- "Introduce EP configurations"
pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/1811