Compare ERNIE 4.5 21B A3B Thinking from baidu and DeepSeek V3.2 from DeepSeek on key metrics including benchmarks, price, context length, and other model features. Access both models and hundreds of others through the OpenRouter API.


ERNIE 4.5 21B A3B Thinking and DeepSeek V3.2 are available through the OpenRouter API, so switching between them takes a model slug change rather than a new integration.
ERNIE 4.5 21B A3B Thinking, from baidu, has a 131,072-token context window and is priced by the provider serving it on OpenRouter.
DeepSeek V3.2, from DeepSeek, has a 163,840-token context window and is priced at $0.2088/M tokens input, $0.3096/M tokens output on OpenRouter.
| ERNIE 4.5 21B A3B Thinking | DeepSeek V3.2 | |
|---|---|---|
| Input price | Pricing varies by provider | $0.2088/M tokens |
| Output price | Pricing varies by provider | $0.3096/M tokens |
| Context window | 131,072 tokens | 163,840 tokens |
| Intelligence Index | Not available | 21.5 |
| Coding Index | Not available | 44.2 |
| Agentic Index | Not available | Not available |
| Latency (p50) | Not available | 1.53 s |
| Throughput (p50) | Not available | 36.0 tok/s |
DeepSeek V3.2 has the longer context window at 163,840 tokens, compared with 131,072 tokens for ERNIE 4.5 21B A3B Thinking.