cp: 26.06 perf summary updates (4384) into r0.5.0 (#4398)

Signed-off-by: Malay Nagda <malayn@nvidia.com>
Signed-off-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
Signed-off-by: NeMo Bot <nemo-bot@nvidia.com>
Co-authored-by: malay-nagda <malayn@nvidia.com>
Co-authored-by: nemo-autobot[bot] <272199896+nemo-autobot[bot]@users.noreply.github.com>
Co-authored-by: Automation Bot <nemo-toolkit@nvidia.com>
Co-authored-by: dimapihtar <37850217+dimapihtar@users.noreply.github.com>

Author: 5 people

docs/performance-summary.md

@@ -62,17 +62,16 @@ The performance data includes:
 
 | System | #-GPUs | Precision | GBS | MBS | Sequence Length | TP | PP | CP | VP | EP | Tokens / sec / GPU | Model TFLOP / sec / GPU |
 |--------|--------|-----------|-----|-----|-----------------|----|----|----|----|----|-----------------------|-------------------------|
-| DGX-GB300 | 64 | BF16 | 1280 | 4 | 4096 | 1 | 1 | 1 | n/a | 64 | 20635 | 673 |
-| DGX-GB200 | 64 | BF16 | 1280 | 4 | 4096 | 1 | 1 | 1 | n/a | 64 | 17770 | 580 |
-| DGX-H100 | 64 | BF16 | 1280 | 1 | 4096 | 1 | 4 | 1 | n/a | 8 | 5860 | 191 |
+| DGX-GB300 | 64 | MXFP8 | 1280 | 4 | 4096 | 1 | 1 | 1 | n/a | 16 | 33166 | 1081 |
+| DGX-GB200 | 64 | MXFP8 | 1280 | 4 | 4096 | 1 | 1 | 1 | n/a | 64 | 28947 | 943 |
 
 #### Model: Qwen3_30B_a3B
 
 | System | #-GPUs | Precision | GBS | MBS | Sequence Length | TP | PP | CP | VP | EP | Tokens / sec / GPU | Model TFLOP / sec / GPU |
 |--------|--------|-----------|-----|-----|-----------------|----|----|----|----|----|-----------------------|-------------------------|
 | DGX-GB300 | 8 | MXFP8 | 512 | 8 | 4096 | 1 | 1 | 1 | n/a | 8 | 45275 | 1041 |
 | DGX-GB200 | 8 | MXFP8 | 512 | 4 | 4096 | 1 | 1 | 1 | n/a | 8 | 40706 | 936 |
-| DGX-H100 | 16 | FP8 | 1024 | 1 | 4096 | 1 | 1 | 1 | n/a | 16 | 8467 | 195 |
+| DGX-H100 | 16 | FP8 | 1024 | 1 | 4096 | 1 | 1 | 1 | n/a | 16 | 8826 | 203 |
 
 #### Model: Qwen3_235B_a22B