Compare/ERNIE 5.0 Thinking Preview vs gpt-oss-20B (high)

ERNIE 5.0 Thinking Previewvsgpt-oss-20B (high)

Side-by-side comparison of pricing, 12 benchmarks, and generation speed.

Baidu

ERNIE 5.0 Thinking Preview

Input
$0/M
Output
$0/M
Speed
TTFT
OpenAI

gpt-oss-20B (high)

Input
$0.05/M
Output
$0.2/M
Speed
256 tok/s
TTFT
0.34s

Winner by Category

Cheaper
ERNIE 5.0 Thinking Preview
Faster (tok/s)
gpt-oss-20B (high)
Lower Latency
ERNIE 5.0 Thinking Preview
Benchmarks (8-3)
ERNIE 5.0 Thinking Preview

Pricing Comparison

MetricERNIE 5.0 Thinking Previewgpt-oss-20B (high)
Input ($/M tokens)$0$0.05
Output ($/M tokens)$0$0.2
Cost for 1M input + 100K output tokens:
ERNIE 5.0 Thinking Preview$0.00
gpt-oss-20B (high)$0.07

Speed Comparison

Output Speed (tokens/s) — higher is better
ERNIE 5.0 Thinking Preview
gpt-oss-20B (high)
256 tok/s
Time to First Token (seconds) — lower is better
ERNIE 5.0 Thinking Preview
gpt-oss-20B (high)
0.34s

Benchmark Comparison

Data from Artificial Analysis API — 12 benchmarks

Intelligence Index
29.124.5
Coding Index
29.218.5
Math Index
85.089.3
GPQA Diamond
77.7%68.8%
MMLU-Pro
83.0%74.8%
LiveCodeBench
81.2%77.7%
AIME 2025
85.0%89.3%
MATH-500
Humanity's Last Exam
12.7%9.8%
SciCode
37.5%34.4%
IFBench
41.4%65.1%
TerminalBench
25.0%10.6%
ERNIE 5.0 Thinking Preview8 wins
3 winsgpt-oss-20B (high)

Frequently Asked Questions

Which is cheaper, ERNIE 5.0 Thinking Preview or gpt-oss-20B (high)?

ERNIE 5.0 Thinking Preview is cheaper overall. Its blended price (3:1 input/output ratio) is $0.00/M tokens vs $0.09/M for gpt-oss-20B (high).

Which model performs better on benchmarks?

ERNIE 5.0 Thinking Preview wins 8 out of 12 benchmarks compared to 3 for gpt-oss-20B (high). See the detailed benchmark chart above for per-category results.

Which is faster for real-time applications?

gpt-oss-20B (high) generates tokens faster at 256 tok/s vs 0 tok/s. ERNIE 5.0 Thinking Preview also has lower time-to-first-token (0.00s vs 0.34s).

When should I use ERNIE 5.0 Thinking Preview vs gpt-oss-20B (high)?

Choose based on your priorities: ERNIE 5.0 Thinking Preview for lower cost, ERNIE 5.0 Thinking Preview for stronger benchmark performance, and gpt-oss-20B (high) for faster generation. For latency-sensitive apps, check the TTFT comparison above.