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.
Model details →Baidu: ERNIE 4.5 21B A3B Thinking vs Qwen: Qwen3 30B A3B Thinking 2507
Baidu: ERNIE 4.5 21B A3B Thinking wins on 2 of 6 axes — pricing and capability skew in its favour for most workloads.
Qwen3-30B-A3B-Thinking-2507 is a 30B parameter Mixture-of-Experts reasoning model optimized for complex tasks requiring extended multi-step thinking. The model is designed specifically for “thinking mode,” where internal reasoning traces ar…
Model details →Side-by-side comparison
| Capability | Baidu: ERNIE 4.5 21B A3B Thinking | Qwen: Qwen3 30B A3B Thinking 2507 | Winner |
|---|---|---|---|
| Context window Maximum number of input tokens the model can attend to in a single request. | 131K | 131K | Tie |
| Input price (per 1M) Cost per million input tokens billed by the provider. | $0.07 | $0.08 | 🏆 Baidu: ERNIE 4.5 21B A3B Thinking |
| Output price (per 1M) Cost per million output tokens billed by the provider. | $0.28 | $0.40 | 🏆 Baidu: ERNIE 4.5 21B A3B Thinking |
| Tool / function calling First-class support for emitting structured tool calls. | — | Yes | 🏆 Qwen: Qwen3 30B A3B Thinking 2507 |
| Vision input Accepts image inputs alongside text. | — | — | Tie |
| Reasoning mode Internal chain-of-thought / extended-thinking support. | Yes | Yes | Tie |
Frequently asked questions
Is Baidu: ERNIE 4.5 21B A3B Thinking better than Qwen: Qwen3 30B A3B Thinking 2507?
Baidu: ERNIE 4.5 21B A3B Thinking wins on 2 of 6 axes — pricing and capability skew in its favour for most workloads.
What's the price difference between Baidu: ERNIE 4.5 21B A3B Thinking and Qwen: Qwen3 30B A3B Thinking 2507?
Input: $0.07 vs $0.08 per 1M tokens. Output: $0.28 vs $0.40 per 1M tokens.
What context windows do Baidu: ERNIE 4.5 21B A3B Thinking and Qwen: Qwen3 30B A3B Thinking 2507 support?
Baidu: ERNIE 4.5 21B A3B Thinking supports up to 131K tokens. Qwen: Qwen3 30B A3B Thinking 2507 supports up to 131K tokens.
Do both Baidu: ERNIE 4.5 21B A3B Thinking and Qwen: Qwen3 30B A3B Thinking 2507 support tool calling?
Baidu: ERNIE 4.5 21B A3B Thinking: not advertised. Qwen: Qwen3 30B A3B Thinking 2507: yes.