Consolidate Llama model usage in tests (#13094)

This commit is contained in:
Harry Mellor
2025-02-13 22:18:03 -08:00
committed by GitHub
parent 40932d7a05
commit f2b20fe491
22 changed files with 45 additions and 53 deletions
+4 -6
View File
@@ -28,7 +28,7 @@ def setup_servers():
"-m",
"vllm.entrypoints.openai.api_server",
"--model",
"meta-llama/Meta-Llama-3.1-8B-Instruct",
"meta-llama/Llama-3.2-1B-Instruct",
"--port",
"8100",
"--gpu-memory-utilization",
@@ -49,7 +49,7 @@ def setup_servers():
"-m",
"vllm.entrypoints.openai.api_server",
"--model",
"meta-llama/Meta-Llama-3.1-8B-Instruct",
"meta-llama/Llama-3.2-1B-Instruct",
"--port",
"8200",
"--gpu-memory-utilization",
@@ -100,8 +100,7 @@ def test_disaggregated_prefilling(prompt):
response = requests.post("http://localhost:8100/v1/completions",
headers={"Content-Type": "application/json"},
json={
"model":
"meta-llama/Meta-Llama-3.1-8B-Instruct",
"model": "meta-llama/Llama-3.2-1B-Instruct",
"prompt": prompt,
"max_tokens": 1,
"temperature": 0
@@ -112,8 +111,7 @@ def test_disaggregated_prefilling(prompt):
response = requests.post("http://localhost:8200/v1/completions",
headers={"Content-Type": "application/json"},
json={
"model":
"meta-llama/Meta-Llama-3.1-8B-Instruct",
"model": "meta-llama/Llama-3.2-1B-Instruct",
"prompt": prompt,
"max_tokens": 10,
"temperature": 0