Skip to content
This repository was archived by the owner on Apr 20, 2026. It is now read-only.

Grho/feb2 b add >5 TTFT load points #132

Merged
gracehonv merged 4 commits into from
Feb 3, 2026
Merged

Grho/feb2 b add >5 TTFT load points #132

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
773ef88
add nginx frontend containers
Feb 2, 2026
a3e2797
lustre
ishandhanani Feb 2, 2026
568816e
lustre
ishandhanani Feb 2, 2026
a6a2afa
modify points to put back TTFT>5 points
Feb 3, 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
3 changes: 0 additions & 3 deletions recipes/gb300-fp8/1k1k/stp/low-latency.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -8,9 +8,6 @@ model:
frontend:
nginx_container: nginx

extra_mount: # add this if you need to mount extra directories to the container
- "/lustre:/lustre"

resources:
gpu_type: "gb300"
prefill_nodes: 1
Expand Down
2 changes: 0 additions & 2 deletions recipes/gb300-fp8/1k1k/stp/max.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -10,8 +10,6 @@ model:
frontend:
nginx_container: nginx

extra_mount: # add this if you need to mount extra directories to the container
- "/lustre:/lustre"
resources:
gpu_type: "gb300"
prefill_nodes: 2
Expand Down
2 changes: 0 additions & 2 deletions recipes/gb300-fp8/1k1k/stp/mid.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9,8 +9,6 @@ model:
frontend:
nginx_container: nginx

extra_mount: # add this if you need to mount extra directories to the container
- "/lustre:/lustre"
resources:
gpu_type: "gb300"
prefill_nodes: 4
Expand Down
3 changes: 3 additions & 0 deletions recipes/gb300-fp8/1k8k/stp/low-latency.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -5,6 +5,9 @@ model:
container: "lmsysorg/sglang:v0.5.8-cu130-runtime"
precision: "fp8"

frontend:
nginx_container: nginx

resources:
gpu_type: "gb300"
prefill_nodes: 1
Expand Down
5 changes: 4 additions & 1 deletion recipes/gb300-fp8/1k8k/stp/max.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -7,6 +7,9 @@ model:
container: "lmsysorg/sglang:v0.5.8-cu130-runtime"
precision: "fp8"

frontend:
nginx_container: nginx

resources:
gpu_type: "gb300"
prefill_nodes: 2
Expand Down Expand Up @@ -166,6 +169,6 @@ benchmark:
type: "sa-bench"
isl: 1024
osl: 8192
concurrencies: [8192]
concurrencies: [8192,10240]
req_rate: "inf"

3 changes: 3 additions & 0 deletions recipes/gb300-fp8/1k8k/stp/mid.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -7,6 +7,9 @@ model:
container: "lmsysorg/sglang:v0.5.8-cu130-runtime"
precision: "fp8"

frontend:
nginx_container: nginx

resources:
gpu_type: "gb300"
prefill_nodes: 2
Expand Down
3 changes: 0 additions & 3 deletions recipes/gb300-fp8/8k1k/stp/low-latency.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -8,9 +8,6 @@ model:
frontend:
nginx_container: nginx

extra_mount: # add this if you need to mount extra directories to the container
- "/lustre:/lustre"

resources:
gpu_type: "gb300"
prefill_nodes: 1
Expand Down
4 changes: 1 addition & 3 deletions recipes/gb300-fp8/8k1k/stp/max.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -10,8 +10,6 @@ model:
frontend:
nginx_container: nginx

extra_mount: # add this if you need to mount extra directories to the container
- "/lustre:/lustre"
resources:
gpu_type: "gb300"
prefill_nodes: 12
Expand Down Expand Up @@ -168,6 +166,6 @@ benchmark:
type: "sa-bench"
isl: 8192
osl: 1024
concurrencies: [2048,4096,10240]
concurrencies: [2048,4096]
req_rate: "inf"

4 changes: 1 addition & 3 deletions recipes/gb300-fp8/8k1k/stp/mid.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -10,8 +10,6 @@ model:
frontend:
nginx_container: nginx

extra_mount: # add this if you need to mount extra directories to the container
- "/lustre:/lustre"
resources:
gpu_type: "gb300"
prefill_nodes: 10
Expand Down Expand Up @@ -168,6 +166,6 @@ benchmark:
type: "sa-bench"
isl: 8192
osl: 1024
concurrencies: [128,256,512,1024,2048,4096]
concurrencies: [128,256,512,1024]
req_rate: "inf"

Loading