EngineX-Hygon/sglang
5
0
Fork 0
Code Issues Pull Requests Actions 7 Projects Releases Wiki Activity

Fix a draft model accuracy bug in eagle; support step=1; return logprob in eagle (#4134)

Browse Source 
Co-authored-by: Sehoon Kim <kssteven418@gmail.com>
Co-authored-by: SangBin Cho <rkooo567@gmail.com>
Co-authored-by: Sehoon Kim <sehoon@x.ai>
This commit is contained in:
Lianmin Zheng
2025-03-06 06:13:59 -08:00
committed by GitHub
parent 3a3918121f
commit bc1534ff32
11 changed files with 304 additions and 106 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
test/srt/test_bench_serving.py
View File

@@ -165,7 +165,7 @@ class TestBenchServing(unittest.TestCase):
f'median_e2e_latency_ms : {res["median_e2e_latency_ms"]:.2f} ms\n'
f'accept_length : {res["accept_length"]:.2f} \n'
)
self.assertLess(res["median_e2e_latency_ms"], 1100)
self.assertLess(res["median_e2e_latency_ms"], 900)
self.assertGreater(res["accept_length"], 2.99)
def test_moe_offline_throughput_default(self):