autodeploy_engine_config.yaml 296 Bytes
Newer Older
1
# SPDX-FileCopyrightText: Copyright (c) 2025-2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
2
3
4
5
6
7
8
9
# SPDX-License-Identifier: Apache-2.0

backend: _autodeploy
kv_cache_config:
  enable_partial_reuse: false
  free_gpu_memory_fraction: 0.80
  max_tokens: 8192
compile_backend: torch-cudagraph