Skip to content

docs: Reference Deepseek R1 configs in TRTLLM README #1414

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
rmccorm4 merged 2 commits into from
Jun 7, 2025
Merged

docs: Reference Deepseek R1 configs in TRTLLM README #1414

Changes from 1 commit
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
9a3a12c
docs: Reference Deepseek R1 configs in TRTLLM README
rmccorm4 Jun 6, 2025
fe958a8
tweak wording on customization
rmccorm4 Jun 6, 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
Next Next commit
docs: Reference Deepseek R1 configs in TRTLLM README
  • Loading branch information
@rmccorm4
rmccorm4 committed Jun 6, 2025
commit 9a3a12c354378d807d8a597d4f6125753bf0ed05
7 changes: 7 additions & 0 deletions examples/tensorrt_llm/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -126,6 +126,13 @@ deploy a single Decode worker on one node, and a single Prefill worker on the ot
However, the instance counts, TP sizes, other configs, and responsibilities of each node
can be customized and deployed in similar ways.

For example, to deploy Deepseek R1, you could replace the referenced example
configs (`configs/agg.yaml`, `configs/disagg.yaml`) with corresponding Deepseek R1
example configs (`configs/deepseek_r1/agg.yaml`, `configs/deepseek_r1/disagg.yaml`).
You can find the example Deepseek R1 configs for GB200
[here](configs/deepseek_r1), but the parallelism settings in the configs can be
customized for other hardware configurations.

##### Head Node

Start nats/etcd:
Expand Down
Loading