Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Time to First Token (TTFT) measurement on Context sequences 32K-256K (vLLM 0.18.0/0.20.0)

2.944TTFT (Dense, s)

Qwen3.5-9B (distilled indexer)

1.0251613.9773326.929539.88167Jun 5, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.06
2.9442.7471.07
2026.06
5.9574.6131.29
2026.06
6.8636.0241.14
2026.06
11.789.1731.28
2026.06
17.65212.4521.42
2026.06
17.92810.4911.71
2026.06
50.91526.3221.93