Skip to content

fix: Migrating trtllm examples from 1.0.0rc0 to 1.0.4rc4 #2217

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
tanmayv25 merged 13 commits into from
Jul 31, 2025
Merged

fix: Migrating trtllm examples from 1.0.0rc0 to 1.0.4rc4 #2217

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
13 commits
Select commit Hold shift + click to select a range
6214cdd
Update from rc0 to rc4.
KrishnanPrash Jul 23, 2025
630b7b1
partial rc4 changes
KrishnanPrash Jul 28, 2025
502c8f0
Adding moe_config
KrishnanPrash Jul 28, 2025
4539733
migrating to cuda_config
KrishnanPrash Jul 28, 2025
3540b43
Updating configs
KrishnanPrash Jul 28, 2025
bc3bcf6
Revamping configs: typos, sub-fields, ...
KrishnanPrash Jul 28, 2025
82ffd54
attempt to fix gpu OOM issue
KrishnanPrash Jul 29, 2025
efe02fc
kv_cache_config fix for KV Routing Path
KrishnanPrash Jul 31, 2025
f53beab
Merge remote-tracking branch 'origin/main' into kprashanth/trtllm_rc4
KrishnanPrash Jul 31, 2025
3d8f400
Fixes
KrishnanPrash Jul 31, 2025
01f847e
Config fixes
KrishnanPrash Jul 31, 2025
edc4575
cuda_graph_config change
KrishnanPrash Jul 31, 2025
004a801
Undo extra changes
KrishnanPrash Jul 31, 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
attempt to fix gpu OOM issue
  • Loading branch information
@KrishnanPrash
KrishnanPrash committed Jul 29, 2025
commit 82ffd54aa561b2c9d614c16e462714af26d71ec9
2 changes: 1 addition & 1 deletion components/backends/trtllm/engine_configs/deepseek_r1/wide_ep/wide_ep_agg.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -22,7 +22,7 @@ max_num_tokens: 256
max_seq_len: 8448

kv_cache_config:
free_gpu_memory_fraction: 0.7
free_gpu_memory_fraction: 0.3
dtype: fp8

cuda_graph_config:
Expand Down
2 changes: 1 addition & 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 @@ -30,7 +30,7 @@ max_num_tokens: 8192
max_seq_len: 8192

kv_cache_config:
free_gpu_memory_fraction: 0.75
free_gpu_memory_fraction: 0.3
dtype: fp8 # NOTE: This dtype must match in both prefill/decode configs

# NOTE: pytorch_backend_config section flattened since: https://github.com/NVIDIA/TensorRT-LLM/pull/4603
Expand Down