A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 17, 01:45 UTC
[Misc] fix typo in llm_engine init logging (#4428)
DefTruth committed
Apr 28, 2024 at 18:58 UTC
9c7306ac114da3e31a5ff040a76f6c640354cce8
1 file changed
+1
-1
vllm/engine/llm_engine.py
+1
-1
@@ -101,7 +101,7 @@ class LLMEngine:
101
"skip_tokenizer_init=%s, tokenizer_mode=%s, revision=%s, "
102
"tokenizer_revision=%s, trust_remote_code=%s, dtype=%s, "
103
"max_seq_len=%d, download_dir=%r, load_format=%s, "
104
- "tensor_parallel_size=%d, disable_custom_all_reduce=%s"
104
+ "tensor_parallel_size=%d, disable_custom_all_reduce=%s, "
105
"quantization=%s, enforce_eager=%s, kv_cache_dtype=%s, "
106
"quantization_param_path=%s, device_config=%s, "
107
"decoding_config=%r, seed=%d)",