Model comparison
ERNIE 4.5 21B A3B vs Baidu: Qianfan-OCR-Fast
Compare ERNIE 4.5 21B A3B and Baidu: Qianfan-OCR-Fast: input/output $/Mtoken, context window, modalities, and license. OpenRouter-synced pricing.
Baidu
ERNIE 4.5 21B A3B
Open-weight instruction model for adaptable chat and self-hosted production workloads
Baidu
Baidu: Qianfan-OCR-Fast
OCR model for extracting structured text from documents and screenshots
| Metric | ERNIE 4.5 21B A3B | Baidu: Qianfan-OCR-Fast |
|---|---|---|
| Provider | Baidu | Baidu |
| Context window | 120,000 | 65,536 |
| Input $/Mtok | $0.070 | $0.680 |
| Output $/Mtok | $0.280 | $2.81 |
| Modalities | text | text, vision |
| License | open-weights | closed |
Quick take
On input price, ERNIE 4.5 21B A3B is cheaper at $0.070/Mtok. For context window, ERNIE 4.5 21B A3B leads with 120,000 tokens.
Pick based on your workload: high-volume cheap inference vs long-document / agent loops. Enterprise DNA can wire either model into Omni with evals, secrets, and job orchestration.