llm-benchmarks
githubdrose.io

baidu Provider Benchmarks


Measured over the last 30 days

Models Tracked
13
Avg Tokens / Second
12.49
Avg Time to First Token (ms)
13803.28
Last Updated
Sep 2, 2026

All baidu models

13 tracked · fastest first
ModelAvg Toks/SecMinMaxAvg TTF (ms)
DeepSeek V4 Flash 073129.3013.3044.202320.00
GLM 5.225.5023.5026.801790.00
DeepSeek V4 Flash 073124.8017.7028.302120.00
GLM 5.124.502.8232.302070.00
GLM 5.123.6021.1024.501880.00
DeepSeek V3.221.5017.7024.101380.00
GLM 5.221.3011.7027.501630.00
GLM 5.213.009.4716.201660.00
DeepSeek V4 Flash 042312.909.0616.004830.00
DeepSeek V4 Pro 042311.908.1014.104370.00
Kimi K2.63.612.326.0819100.00
GLM 52.372.172.7826100.00
Kimi K2.62.122.122.1227700.00

Frequently Asked Questions

Which baidu model is fastest?
Based on recent tests, DeepSeek V4 Flash 0731 shows the highest average throughput among tracked baidu models.
How many recent measurements feed this dashboard?
This provider summary aggregates 3539 individual prompts measured across 3272 monitoring runs over the past month.