Skip to content

[https://nvbugs/5512734][fix] Update kv cache config for maverick #7710

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
litaotju merged 1 commit into from
Sep 15, 2025
Merged

[https://nvbugs/5512734][fix] Update kv cache config for maverick #7710

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: 2 additions & 0 deletions tests/integration/defs/accuracy/test_llm_api_pytorch.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -502,13 +502,15 @@ def test_chunked_prefill(self, attn_backend):
(8, 1, 8)],
ids=["tp8", "tp8ep4", "tp8ep8"])
def test_fp8(self, cuda_graph, tp_size, pp_size, ep_size):
kv_cache_config = KvCacheConfig(free_gpu_memory_fraction=0.75)
with LLM(
f"{llm_models_root()}/llama4-models/nvidia/Llama-4-Maverick-17B-128E-Instruct-FP8",
tensor_parallel_size=tp_size,
# Keep this low to avoid warmup OOM in CI
max_seq_len=8192,
pipeline_parallel_size=pp_size,
moe_expert_parallel_size=ep_size,
kv_cache_config=kv_cache_config,
cuda_graph_config=CudaGraphConfig()
if cuda_graph else None) as llm:
assert llm.args.quant_config.quant_algo == QuantAlgo.FP8
Expand Down