Skip to content

GitLab

Unverified Commit 6acb7a62 authored by Rabi Mishra's avatar Rabi Mishra Committed by GitHub
Browse files

[Misc]Fix benchmarks/README.md for speculative decoding (#18897) 


Signed-off-by: default avatarrabi <ramishra@redhat.com>
parent 4f4a6b84
Hide whitespace changes
Inline Side-by-side
Showing with 6 additions and 6 deletions
benchmarks/README.md
View file @ 6acb7a62
......@@ -146,9 +146,9 @@ python3 vllm/benchmarks/benchmark_serving.py \
``` bash
VLLM_USE_V1=1 vllm serve meta-llama/Meta-Llama-3-8B-Instruct \
--ngram_prompt_lookup_min 2 \
--ngram-prompt-lookup-max 5 \
--speculative_config '{"model": "[ngram]", "num_speculative_tokens": 5}
--speculative-config $'{"method": "ngram",
"num_speculative_tokens": 5, "prompt_lookup_max": 5,
"prompt_lookup_min": 2}'
```
``` bash
......@@ -273,9 +273,9 @@ python3 vllm/benchmarks/benchmark_throughput.py \
--output-len=100 \
--num-prompts=2048 \
--async-engine \
--ngram_prompt_lookup_min=2 \
--ngram-prompt-lookup-max=5 \
--speculative_config '{"model": "[ngram]", "num_speculative_tokens": 5}
--speculative-config $'{"method": "ngram",
"num_speculative_tokens": 5, "prompt_lookup_max": 5,
"prompt_lookup_min": 2}'
```
```
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment