Skip to content

[NVIDIA] fix: gptoss h100 docker bug#310

Merged
cquil11 merged 1 commit intomainfrom
h100-docker-fix
Dec 8, 2025
Merged

[NVIDIA] fix: gptoss h100 docker bug#310
cquil11 merged 1 commit intomainfrom
h100-docker-fix

Conversation

@cquil11
Copy link
Copy Markdown
Collaborator

@cquil11 cquil11 commented Dec 8, 2025

A bug was introduced in #227 where benchmarks/gptoss_fp4_h100_docker.sh has --max-concurrency hard-coded to 512 instead of $CONC. This fixes that regression.

Smoke test: https://github.com/InferenceMAX/InferenceMAX/actions/runs/20041264142

@cquil11 cquil11 requested a review from a team as a code owner December 8, 2025 20:05
@cquil11 cquil11 self-assigned this Dec 8, 2025
@cquil11 cquil11 merged commit 4e135fd into main Dec 8, 2025
19 checks passed
@cquil11 cquil11 deleted the h100-docker-fix branch December 8, 2025 22:20
Oseltamivir pushed a commit that referenced this pull request Dec 9, 2025
@cquil11 cquil11 added the NVIDIA label Apr 8, 2026
@cquil11 cquil11 changed the title fix: gptoss h100 docker bug [NVIDIA] fix: gptoss h100 docker bug Apr 8, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

Projects

Development

Successfully merging this pull request may close these issues.

1 participant