Skip to content
Merged
Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
36 commits
Select commit Hold shift + click to select a range
1aa8c32
chore: update CODEOWNERS for multimodal examples (#2878)
biswapanda Sep 5, 2025
ece24b1
first attempt
tedzhouhk Sep 8, 2025
c5633e1
add aiofiles
tedzhouhk Sep 9, 2025
0846ae7
fix trtllm docker file
tedzhouhk Sep 9, 2025
90b5ff1
Revert "fix trtllm docker file"
tedzhouhk Sep 9, 2025
f2a66cb
install requirement.txt
tedzhouhk Sep 9, 2025
4001324
add trtllm to sla planner
tedzhouhk Sep 9, 2025
ebc7611
fix: fix hermes tool call config (#2915)
ayushag-nv Sep 8, 2025
e63ec2e
ci: OPS-724: Move to ARC runners (#2904)
dillon-cullinan Sep 8, 2025
37f2778
fix: CI is broken with a deprecated dependency on pynvml (#2926)
saturley-hall Sep 8, 2025
cd1115a
fix: fix typo in multinode example (#2931)
julienmancuso Sep 8, 2025
327f3fe
ci: Add concurrency check to auto cancel running actions. (#2438)
pvijayakrish Sep 8, 2025
8b1b24c
chore: added utility to detect possible tool call start for a chunk (…
ayushag-nv Sep 8, 2025
d34cfdd
chore: add preference logic for using tool-call and reasoning parsers…
ayushag-nv Sep 8, 2025
dad62a5
Update README.md (#2938)
harryskim Sep 8, 2025
64ba7f3
build: OPS-597: restructure sglang to follow container strategy struc…
nv-tusharma Sep 8, 2025
766d5b2
refactor: standardize e2e tests across 3 frameworks (#2827)
alec-flowers Sep 8, 2025
e41c5bb
feat: automatically setup and inject prometheus configuration (#2912)
julienmancuso Sep 9, 2025
1803db8
fix: WAR DeepGemm JIT compilation errors (#2937)
GuanLuo Sep 9, 2025
a76fd70
ci: sglang functional tests (#2943)
alec-flowers Sep 9, 2025
8f1f965
feat: update benchmarking and deploy utils (#2933)
hhzhang16 Sep 9, 2025
4db7fcf
feat: Add a checksum to ModelDeploymentCard fields (#2934)
grahamking Sep 9, 2025
351464b
ci: Fix Dockerfile mount secrets (#2960)
dillon-cullinan Sep 9, 2025
51c75e1
chore: added tool call schema validation in oai formatter (#2935)
ayushag-nv Sep 9, 2025
f7090a3
test: remove nighlty marker in kvbm tests (#2958)
nv-anants Sep 9, 2025
f5644ef
ci: remove pre-merge ignore in github workflow (#2940)
nv-anants Sep 9, 2025
1a412eb
ci: longer timeout, change model for l40 (#2951)
alec-flowers Sep 9, 2025
b19deaf
fix: aggregate logprobs (#2928)
messiaen Sep 9, 2025
7148426
fix: no reasoning parser by default (#2939)
nealvaidya Sep 9, 2025
a2e3b52
docs: fix broken links (#2965)
nv-nmailhot Sep 9, 2025
37213b6
feat: add a virtual connector for 3rd party deployments (#2913)
tedzhouhk Sep 9, 2025
3af3425
fix: dyn namespace scoping for trtllm
biswapanda Sep 9, 2025
436307c
pc
tedzhouhk Sep 10, 2025
e62f664
remove duplicate
tedzhouhk Sep 10, 2025
a4a3e66
Merge branch 'main' of https://github.com/ai-dynamo/dynamo into hzhou…
tedzhouhk Sep 10, 2025
e6ac2a7
address pr comment
tedzhouhk Sep 10, 2025
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
Prev Previous commit
Next Next commit
Merge branch 'main' of https://github.com/ai-dynamo/dynamo into hzhou…
…/trtllm-sweep

Signed-off-by: hongkuanz <[email protected]>
  • Loading branch information
tedzhouhk committed Sep 10, 2025
commit a4a3e66bcb75de64995ae801bad4f408eec98116
2 changes: 1 addition & 1 deletion container/Dockerfile.sglang
Original file line number Diff line number Diff line change
Expand Up @@ -164,7 +164,7 @@ RUN ARCH=$(dpkg --print-architecture) && \
arm64) PLATFORM=linux-arm64 ;; \
*) echo "Unsupported architecture: $ARCH" && exit 1 ;; \
esac && \
curl -fsSL "https://github.com/prometheus/prometheus/releases/download/v${PROM_VERSION}/prometheus-${PROM_VERSION}.${PLATFORM}.tar.gz" \
curl -fsSL --retry 5 --retry-delay 5 "https://github.com/prometheus/prometheus/releases/download/v${PROM_VERSION}/prometheus-${PROM_VERSION}.${PLATFORM}.tar.gz" \
| tar -xz -C /tmp && \
mv "/tmp/prometheus-${PROM_VERSION}.${PLATFORM}/prometheus" /usr/local/bin/ && \
chmod +x /usr/local/bin/prometheus && \
Expand Down
10 changes: 5 additions & 5 deletions docs/benchmarks/pre_deployment_profiling.md
Original file line number Diff line number Diff line change
Expand Up @@ -127,13 +127,13 @@ Use the default pre-built image and inject custom configurations via PVC:
2. **Inject your custom disagg configuration:**
```bash
# Use default disagg.yaml config
python3 deploy/utils/inject_manifest.py --namespace $NAMESPACE --src components/backends/vllm/deploy/disagg.yaml --dest /data/configs/disagg.yaml
python3 -m deploy.utils.inject_manifest --namespace $NAMESPACE --src components/backends/vllm/deploy/disagg.yaml --dest /data/configs/disagg.yaml

# Or use a custom disagg config file
python3 deploy/utils/inject_manifest.py --namespace $NAMESPACE --src my-custom-disagg.yaml --dest /data/configs/disagg.yaml
python3 -m deploy.utils.inject_manifest --namespace $NAMESPACE --src my-custom-disagg.yaml --dest /data/configs/disagg.yaml

# Or specify a custom target path in the PVC
python3 deploy/utils/inject_manifest.py --namespace $NAMESPACE --src my-custom-disagg.yaml --dest /data/profiling_results/my-disagg.yaml
python3 -m deploy.utils.inject_manifest --namespace $NAMESPACE --src my-custom-disagg.yaml --dest /data/profiling_results/my-disagg.yaml
```

> **Note**: All paths must start with `/data/` for security reasons. If you forget this prefix, the script will show a helpful error message with the correct path.
Expand Down Expand Up @@ -193,10 +193,10 @@ To download the results:

```bash
# Download to directory
python3 deploy/utils/download_pvc_results.py --namespace $NAMESPACE --output-dir ./results --folder /data/profiling_results
python3 -m deploy.utils.download_pvc_results --namespace $NAMESPACE --output-dir ./results --folder /data/profiling_results

# Download without any of the auto-created config.yaml files used in profiling
python3 deploy/utils/download_pvc_results.py --namespace $NAMESPACE --output-dir ./results --folder /data/profiling_results --no-config
python3 -m deploy.utils.download_pvc_results --namespace $NAMESPACE --output-dir ./results --folder /data/profiling_results --no-config
```

The script will:
Expand Down
Loading
You are viewing a condensed version of this merge commit. You can view the full changes here.