Skip to content

fix: Fix message truncation in disagg flow #1572

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
merged 1 commit into from
Jun 17, 2025
Merged

fix: Fix message truncation in disagg flow #1572

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
8 changes: 2 additions & 6 deletions examples/tensorrt_llm/configs/llmapi_disagg_configs/single_node_config.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -30,9 +30,7 @@ context_servers:
max_batch_size: 16
enable_chunked_prefill: false
kv_cache_config:
free_gpu_memory_fraction: 0.40
cache_transceiver_config:
max_num_tokens: 10240
free_gpu_memory_fraction: 0.75
# NOTE: pytorch_backend_config section flattened since: https://github.com/NVIDIA/TensorRT-LLM/pull/4603
# NOTE: This field is called 'enable_overlap_scheduler' in older TRTLLM versions
# Overlap scheduler not currently supported in context-only
Expand All @@ -47,9 +45,7 @@ generation_servers:
max_num_tokens: 256
max_batch_size: 256
kv_cache_config:
free_gpu_memory_fraction: 0.40
cache_transceiver_config:
max_num_tokens: 256
free_gpu_memory_fraction: 0.75
# NOTE: pytorch_backend_config section flattened since: https://github.com/NVIDIA/TensorRT-LLM/pull/4603
# NOTE: This field is called 'enable_overlap_scheduler' in older TRTLLM versions
disable_overlap_scheduler: false
Expand Down
8 changes: 2 additions & 6 deletions examples/tensorrt_llm/configs/llmapi_disagg_router_configs/single_node_config.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -30,11 +30,9 @@ context_servers:
max_batch_size: 16
enable_chunked_prefill: false
kv_cache_config:
free_gpu_memory_fraction: 0.40
free_gpu_memory_fraction: 0.75
event_buffer_max_size: 1024
enable_block_reuse: true
cache_transceiver_config:
max_num_tokens: 10240
# NOTE: pytorch_backend_config section flattened since: https://github.com/NVIDIA/TensorRT-LLM/pull/4603
# NOTE: This field is called 'enable_overlap_scheduler' in older TRTLLM versions
# Overlap scheduler not currently supported in context-only
Expand All @@ -50,11 +48,9 @@ generation_servers:
max_num_tokens: 256
max_batch_size: 256
kv_cache_config:
free_gpu_memory_fraction: 0.40
free_gpu_memory_fraction: 0.75
event_buffer_max_size: 1024
enable_block_reuse: true
cache_transceiver_config:
max_num_tokens: 256
# NOTE: pytorch_backend_config section flattened since: https://github.com/NVIDIA/TensorRT-LLM/pull/4603
# NOTE: This field is called 'enable_overlap_scheduler' in older TRTLLM versions
disable_overlap_scheduler: false
Expand Down
Loading