Baidu: ERNIE 4.5 21B A3B Thinking
Analysis Summary
Baidu: ERNIE 4.5 21B A3B Thinking sits in the Legacy tier on our leaderboard, ranked #345 of 525 published models on overall intelligence. At $0.070 input and $0.280 output per 1M tokens, it is among the most expensive on the market. It offers a standard large context window and supports reasoning.
Editorial notes
Baidu's ERNIE 4.5 21B A3B Thinking is a very low-cost model with no published benchmark data, making performance assessment speculative. Without verifiable results and with limited Western business support, it is not recommended for professional deployment at this stage.
Assessed April 23, 2026
Rankings consider pricing, capabilities, benchmarks, and real-world applicability and are refreshed as new models launch. Feedback?
Performance Profile
ERNIE-4.5-21B-A3B-Thinking is Baidu's upgraded lightweight MoE model, refined to boost reasoning depth and quality for top-tier performance in logical puzzles, math, science, coding, text generation, and expert-level academic benchmarks.
How does Baidu: ERNIE 4.5 21B A3B Thinking stack up?
Compare side-by-side with other legacy models.
Model Information
| OpenRouter ID |
baidu/ernie-4.5-21b-a3b-thinking
|
| Provider | baidu |
| Release Date | October 9, 2025 |
| Context Length | 131,072 tokens |
| Max Completion | 65,536 tokens |
| Status | Active |
Pricing
| Token Type | Cost per 1M tokens | Cost per 1K tokens |
|---|---|---|
| Input | $0.07 | $0.000070 |
| Output | $0.28 | $0.000280 |
External Resources
Explore Related Models
Data sourced from OpenRouter API, Artificial Analysis and Hugging Face Open LLM Leaderboard. Scores are editorially curated by our team.
Last updated: April 25, 2026 8:38 pm