Skip to content

perf(agentx): Add B300 vLLM MTP TP4 data points #2647

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
cquil11 merged 3 commits into from
Aug 21, 2026
Merged

perf(agentx): Add B300 vLLM MTP TP4 data points #2647

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
aafd99e
perf(agentx): append B300 TP4 points
ivanium Aug 19, 2026
509c65f
Merge branch 'main' into agent/update-b300-agentx-vllm-mtp-image
cquil11 Aug 21, 2026
efac166
Update perf-changelog.yaml
cquil11 Aug 21, 2026
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
2 changes: 1 addition & 1 deletion configs/nvidia-master.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1734,7 +1734,7 @@ dsv4-fp4-b300-vllm-agentic-mtp:
# TP8 GPU-resident + MTP (num_speculative_tokens=3)
- { tp: 8, kv-offloading: none, spec-decoding: mtp, conc-list: [1, 4] }
# TP4 GPU-resident + MTP (num_speculative_tokens=3)
- { tp: 4, kv-offloading: none, spec-decoding: mtp, conc-list: [8] }
- { tp: 4, kv-offloading: none, spec-decoding: mtp, conc-list: [1, 2, 4, 6, 8] }
# TP4 SimpleCPU + MTP (num_speculative_tokens=3)
- { tp: 4, kv-offloading: dram, kv-offload-backend: { name: vllm-simple }, spec-decoding: mtp, conc-list: [16] }
# DEP4 SimpleCPU + MTP (num_speculative_tokens=3)
Expand Down
10 changes: 10 additions & 0 deletions perf-changelog.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6307,3 +6307,13 @@
- "Disable TensorRT-LLM return_perf_metrics in the prefill and decode workers."
- "Route the selected GB300 workloads to batch_3."
pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/2690

- config-keys:
- dsv4-fp4-b300-vllm-agentic-mtp
scenario-type:
- agentic-coding
description:
- "Append GPU-resident TP4 concurrency 1, 2, 4, and 6 points to the existing B300 DeepSeek-V4-Pro vLLM AgentX MTP curve."
pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/2647
append-only: true