Skip to content

fix: Update disagg configs for trtllm 1.0.0rc4 changes (main) #2276

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

Closed
wants to merge 1 commit into from
Closed

fix: Update disagg configs for trtllm 1.0.0rc4 changes (main) #2276

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
2 changes: 1 addition & 1 deletion components/backends/trtllm/engine_configs/decode.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -28,4 +28,4 @@ kv_cache_config:
free_gpu_memory_fraction: 0.95

cache_transceiver_config:
backend: default
backend: default
5 changes: 4 additions & 1 deletion components/backends/trtllm/engine_configs/deepseek_r1/mtp/mtp_decode.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -51,4 +51,7 @@ cuda_graph_config:
- 128
- 256

print_iter_log: true
print_iter_log: true

cache_transceiver_config:
backend: default
3 changes: 3 additions & 0 deletions components/backends/trtllm/engine_configs/deepseek_r1/mtp/mtp_prefill.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -36,3 +36,6 @@ disable_overlap_scheduler: true
speculative_config:
decoding_type: MTP
num_nextn_predict_layers: 1

cache_transceiver_config:
backend: default
3 changes: 3 additions & 0 deletions components/backends/trtllm/engine_configs/deepseek_r1/simple/decode.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -55,3 +55,6 @@ cuda_graph_config:
- 256

print_iter_log: true

cache_transceiver_config:
backend: default
5 changes: 4 additions & 1 deletion components/backends/trtllm/engine_configs/deepseek_r1/simple/prefill.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -33,4 +33,7 @@ kv_cache_config:
# config field from 'enable_overlap_scheduler' to 'disable_overlap_scheduler':
# https://github.com/NVIDIA/TensorRT-LLM/commit/b4e5df0ee0024eda3eeb83a6ba822245a30ab428
disable_overlap_scheduler: true
print_iter_log: true
print_iter_log: true

cache_transceiver_config:
backend: default
3 changes: 3 additions & 0 deletions components/backends/trtllm/engine_configs/deepseek_r1/wide_ep/wide_ep_decode.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -61,3 +61,6 @@ cuda_graph_config:


print_iter_log: true

cache_transceiver_config:
backend: default
5 changes: 4 additions & 1 deletion components/backends/trtllm/engine_configs/deepseek_r1/wide_ep/wide_ep_prefill.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -38,4 +38,7 @@ kv_cache_config:
# config field from 'enable_overlap_scheduler' to 'disable_overlap_scheduler':
# https://github.com/NVIDIA/TensorRT-LLM/commit/b4e5df0ee0024eda3eeb83a6ba822245a30ab428
disable_overlap_scheduler: true
print_iter_log: true
print_iter_log: true

cache_transceiver_config:
backend: default
3 changes: 3 additions & 0 deletions components/backends/trtllm/engine_configs/llama4/eagle/eagle_decode.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -49,3 +49,6 @@ cuda_graph_config:
- 256

print_iter_log: true

cache_transceiver_config:
backend: default
3 changes: 3 additions & 0 deletions components/backends/trtllm/engine_configs/llama4/eagle/eagle_prefill.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -34,3 +34,6 @@ speculative_config:
kv_cache_config:
free_gpu_memory_fraction: 0.5
enable_block_reuse: false

cache_transceiver_config:
backend: default
3 changes: 3 additions & 0 deletions components/backends/trtllm/engine_configs/llama4/eagle_one_model/eagle_decode.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -38,3 +38,6 @@ cuda_graph_config:
max_batch_size: 256

print_iter_log: true

cache_transceiver_config:
backend: default
3 changes: 3 additions & 0 deletions components/backends/trtllm/engine_configs/llama4/eagle_one_model/eagle_prefill.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -32,3 +32,6 @@ speculative_config:
kv_cache_config:
free_gpu_memory_fraction: 0.5
enable_block_reuse: false

cache_transceiver_config:
backend: default
Loading