Claude Sonnet 4 vs ERNIE 5.0 Thinking Preview
Wins 4 of 7 areas
Coding · Facts · Long documents · Following instructions
Wins 3 of 7 areas
Agents · Reasoning · Images and charts
Claude Sonnet 4 wins more areas, narrowly.ERNIE 5.0 Thinking Preview is cheaper and better at reasoning.
Scores updated · 19 tests both models report · How we compare
Where each one wins
Tests won in each of the seven areas where both have results. Each piece is one test, so a longer bar means more evidence; grey means the two scored within a point of each other.
- CodingWriting and fixing software20Claude Sonnet 42 of 2 tests
- FactsGetting facts right instead of making them up10Claude Sonnet 41 of 2 tests · 1 tie
- Long documentsFinding answers in very long texts10Claude Sonnet 41 of 1 test
- Following instructionsDoing exactly what it is asked10Claude Sonnet 41 of 1 test
- ReasoningHard problems that need careful thinking02ERNIE 5.0 Thinking Preview2 of 3 tests · 1 tie
- AgentsCarrying out multi-step tasks on its own01ERNIE 5.0 Thinking Preview1 of 1 test
- Images and chartsUnderstanding pictures, charts and video01ERNIE 5.0 Thinking Preview1 of 1 test
Agents, images and charts, long documents and following instructions rest on a single test each.
What it costs
Prices per million tokens, roughly 750,000 words. The bars show the cost of a million tokens read plus a million written.
ERNIE 5.0 Thinking Preview costs 75% less for the same work.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where Claude Sonnet 4 pulls ahead
- Reasons across sets of long documentsAA-LCR+63.3points ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+57.9points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+13.3points ahead
Where ERNIE 5.0 Thinking Preview pulls ahead
- Real work tasks from 44 professionsGDPVal+4.7points ahead
- Harder college exam questions with imagesMMMU-Pro+2.8points ahead
- Very hard expert questions across many subjectsHumanity's Last Exam+2.6points ahead
Every test, side by side
All 19 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingClaude Sonnet 4
- Terminal-Bench HardClaude Sonnet 4 by 6.131.125+6.1
- SciCodeClaude Sonnet 4 by 2.54037.5+2.5
AgentsERNIE 5.0 Thinking Preview
- GDPValERNIE 5.0 Thinking Preview by 4.79.714.4+4.7
ReasoningERNIE 5.0 Thinking Preview
- Humanity's Last ExamERNIE 5.0 Thinking Preview by 2.610.713.3+2.6
- CritPtERNIE 5.0 Thinking Preview by 1.10.31.4+1.1
- GPQA Diamondtie77.777.7tie
FactsClaude Sonnet 4
- AA-Omniscience · Non-hallucinationClaude Sonnet 4 by 57.970.913+57.9
- AA-Omniscience · Accuracytie22.722.1tie
Images and chartsERNIE 5.0 Thinking Preview
- MMMU-ProERNIE 5.0 Thinking Preview by 2.861.864.6+2.8
Following instructionsClaude Sonnet 4
- IFBenchClaude Sonnet 4 by 13.354.741.4+13.3
Other results8 tests, not counted
Tests outside the eight areas. They are not counted above: several are summary scores built from other tests, or the same test under another name.
- SimpleQAERNIE 5.0 Thinking Preview by 59.115.975+59.1
- AA-OmniscienceClaude Sonnet 4 by 45.80.2-45.6+45.8
- AA Agentic IndexERNIE 5.0 Thinking Preview by 22.117.639.7+22.1
- τ²-Bench Telecom (AA run)ERNIE 5.0 Thinking Preview by 19.364.683.9+19.3
- AIME 2025ERNIE 5.0 Thinking Preview by 16.570.587+16.5
- Artificial Analysis Coding IndexClaude Sonnet 4 by 8.437.629.2+8.4
- AA IntelligenceClaude Sonnet 4 by 4.618.914.3+4.6
- MMLU-ProERNIE 5.0 Thinking Preview by 3.383.787+3.3
Questions people ask
Which is better, Claude Sonnet 4 or ERNIE 5.0 Thinking Preview?
Claude Sonnet 4 wins four of the seven areas where both have results: coding, facts, long documents and following instructions. ERNIE 5.0 Thinking Preview wins agents, reasoning and images and charts, and costs 75% less.
Which is better for coding?
Claude Sonnet 4. It wins 2 of the 2 coding tests both models report; ERNIE 5.0 Thinking Preview wins none.
Which is cheaper?
Claude Sonnet 4 costs $3.00 per million input tokens and $15.00 per million output tokens; ERNIE 5.0 Thinking Preview costs $0.89 and $3.58. That makes ERNIE 5.0 Thinking Preview about 75% cheaper for the same work.
How do you compare the two?
We use the 19 benchmark tests both models have published scores on. The verdict counts the 11 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 8 are listed but not counted, because several are summary scores or repeat a test. Each score is the one shown on the model's own page.