Compare · ModelsLive · 2 picked · head to head
Qwen3 30B A3B Instruct 2507 vs ERNIE 4.5 21B A3B Thinking
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Qwen3 30B A3B Instruct 2507 wins on 4/6 benchmarks
Qwen3 30B A3B Instruct 2507 wins 4 of 6 shared benchmarks. Leads in knowledge · language.
Category leads
math·ERNIE 4.5 21B A3B Thinkingknowledge·Qwen3 30B A3B Instruct 2507language·Qwen3 30B A3B Instruct 2507coding·ERNIE 4.5 21B A3B Thinking
Hype vs Reality
Attention vs performance
Qwen3 30B A3B Instruct 2507
#68 by perf·no signal
ERNIE 4.5 21B A3B Thinking
#57 by perf·no signal
Best value
ERNIE 4.5 21B A3B Thinking
1.1x better value than Qwen3 30B A3B Instruct 2507
Qwen3 30B A3B Instruct 2507
283.6 pts/$
$0.20/M
ERNIE 4.5 21B A3B Thinking
325.7 pts/$
$0.18/M
Vendor risk
Who is behind the model
Alibaba (Qwen)
$293.0B·Tier 1
Baidu
$45.0B·Tier 1
Head to head
6 benchmarks · 2 models
Qwen3 30B A3B Instruct 2507ERNIE 4.5 21B A3B Thinking
OpenCompass · AIME2025
ERNIE 4.5 21B A3B Thinking leads by +12.4
Qwen3 30B A3B Instruct 2507
63.8
ERNIE 4.5 21B A3B Thinking
76.2
OpenCompass · GPQA-Diamond
Qwen3 30B A3B Instruct 2507 leads by +3.4
Qwen3 30B A3B Instruct 2507
63.6
ERNIE 4.5 21B A3B Thinking
60.2
OpenCompass · HLE
Qwen3 30B A3B Instruct 2507 leads by +1.2
Qwen3 30B A3B Instruct 2507
7.7
ERNIE 4.5 21B A3B Thinking
6.5
OpenCompass · IFEval
Qwen3 30B A3B Instruct 2507 leads by +2.7
Qwen3 30B A3B Instruct 2507
83.9
ERNIE 4.5 21B A3B Thinking
81.2
OpenCompass · LiveCodeBenchV6
ERNIE 4.5 21B A3B Thinking leads by +8.3
Qwen3 30B A3B Instruct 2507
38.7
ERNIE 4.5 21B A3B Thinking
47.0
OpenCompass · MMLU-Pro
Qwen3 30B A3B Instruct 2507 leads by +3.1
Qwen3 30B A3B Instruct 2507
73.9
ERNIE 4.5 21B A3B Thinking
70.8
Full benchmark table
| Benchmark | Qwen3 30B A3B Instruct 2507 | ERNIE 4.5 21B A3B Thinking |
|---|---|---|
OpenCompass · AIME2025 | 63.8 | 76.2 |
OpenCompass · GPQA-Diamond | 63.6 | 60.2 |
OpenCompass · HLE | 7.7 | 6.5 |
OpenCompass · IFEval | 83.9 | 81.2 |
OpenCompass · LiveCodeBenchV6 | 38.7 | 47.0 |
OpenCompass · MMLU-Pro | 73.9 | 70.8 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| $0.09 | $0.30 | 262K tokens (~131 books) | $1.43 | |
| $0.07 | $0.28 | 131K tokens (~66 books) | $1.23 |