Baidu — pricing & performance
Models served
13
of 1,191 tracked models
Mean output speed
68
tokens per second, averaged over
13 measured endpoints
13 measured endpoints
Mean latency
1.1s
time to first token, averaged over
13 measured endpoints
13 measured endpoints
Cheapest blended
$0.023
per 1M tokens at a 3:1
input:output blend
input:output blend
Output speed on Baidu
Output tokens per second for all 13 models on Baidu with published throughput telemetry, the most recent figure OpenRouter publishes for each host. Higher is better. We do not run these measurements ourselves.
The Known Good
13 models on Baidu
Full leaderboard →
| Model | Creator | Input $/M | Output $/M | Blended | Speed | TTFT | Context |
|---|---|---|---|---|---|---|---|
DeepSeek: DeepSeek V3.2
|
DeepSeek | $0.28 | $0.42 | $0.32 | 39 | 1.1s | 131k |
|
DeepSeek: DeepSeek V4 Flash 0423
|
DeepSeek | $0.018 | $0.036 | $0.023 | 65 | 0.9s | 1.05M |
|
DeepSeek: DeepSeek V4 Flash 0731
|
DeepSeek | $0.44 | $1.32 | $0.66 | 115 | 0.7s | 1.05M |
|
DeepSeek: DeepSeek V4 Pro 0423
|
DeepSeek | $0.29 | $0.58 | $0.36 | 58 | 0.9s | 1.05M |
|
DeepSeek: DeepSeek V4 Pro 0813
|
DeepSeek | $0.14 | $0.42 | $0.21 | 46 | 1.3s | 1.05M |
|
DeepSeek: DeepSeek V4.1 Flash
|
DeepSeek | $0.035 | $0.14 | $0.062 | 126 | 1.1s | 1.05M |
MoonshotAI: Kimi K2.6
|
Moonshot AI | $0.43 | $1.83 | $0.78 | 49 | 3.2s | 262k |
Tencent: Hy3
|
Tencent | $0.13 | $0.53 | $0.23 | 65 | 0.9s | 262k |
|
Tencent: Hy3 preview
|
Tencent | $0.066 | $0.26 | $0.11 | 51 | 0.5s | 262k |
Z.ai: GLM 5
|
Z.ai | $0.70 | $2.24 | $1.08 | 65 | 1.2s | 203k |
|
Z.ai: GLM 5.1
|
Z.ai | $0.96 | $3.03 | $1.48 | 63 | 1.4s | 203k |
|
Z.ai: GLM 5.2
|
Z.ai | $0.19 | $0.59 | $0.29 | 78 | 0.9s | 1.05M |
|
Z.ai: GLM 5.3
|
Z.ai | $1.12 | $3.52 | $1.72 | 67 | 0.7s | 1.05M |
The Known Good
Source. Prices, endpoint availability and throughput for Baidu are ingested from OpenRouter;
model metadata is enriched from models.dev. A blank cell means the figure was not published for that endpoint —
it never means zero. Blended price is (3 × input + output) ÷ 4, the same 3:1 blend used everywhere on this
site; see methodology. Quality scores are not provider-specific:
a model scores the same wherever it is hosted, so quality lives on the
model pages.