# SPDX-FileCopyrightText: Copyright (c) 2025-2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved. # SPDX-License-Identifier: Apache-2.0 backend: _autodeploy kv_cache_config: enable_partial_reuse: false free_gpu_memory_fraction: 0.80 max_tokens: 8192 attn_backend: trtllm max_batch_size: 128