ERNIE 5.0 Thinking Preview

ERNIE 5.0 Thinking Preview

Baidu · Released Nov 13, 2025 Efficient
Intelligence #187 / 525
41.3 Our Score
AA Index #100 / 353
29.1 Artificial Analysis
Input
Not priced
Output
Not priced
Context
Not reported

Analysis Summary

ERNIE 5.0 Thinking Preview sits in the Efficient tier on our leaderboard, ranked #187 of 525 published models on overall intelligence. At $0.000 input and $0.000 output per 1M tokens, it is among the most expensive on the market.

Editorial notes

ERNIE 5.0 Thinking Preview from Baidu delivers strong math, livecodebench, and agentic scores with a 29-point intelligence index, though no pricing data is available and it carries a -4 regional accessibility adjustment.

Assessed April 26, 2026

Rankings consider pricing, capabilities, benchmarks, and real-world applicability and are refreshed as new models launch. Feedback?

Performance Profile

Intelligence5.6Technical6Value0Content4.5
Intelligence 5.6/10
Technical 6/10
Content 4.5/10
Value 0/10

Performance Indices

Source: Artificial Analysis

29.1 Intelligence Index
29.2 Coding Index
54.4 Agentic Index
85 Math Index

Benchmark Scores

Intelligence

GPQA Diamond 77.7% Graduate-level scientific reasoning
HLE 12.7% Humanity's Last Exam
MMLU Pro 83% Multi-task language understanding
AIME 2025 85% Competition mathematics (2025)
SciCode 37.5% Scientific computing

Technical

LiveCodeBench 81.2% Live coding evaluation
TerminalBench Hard 25% Agentic terminal tasks
τ²-Bench 83.9% Conversational agent benchmark

Content

IFBench 41.4% Instruction following
LCR 6.7% Long-context reasoning

Benchmark data from Artificial Analysis and Hugging Face

How does ERNIE 5.0 Thinking Preview stack up?

Compare side-by-side with other efficient models.

Compare Models

Model Information

ProviderBaidu
Release Date November 13, 2025
Status Active