Claude Sonnet 4 vs EXAONE 4.5 33B
Wins 4 of 7 areas
Coding · Agents · Facts · Long documents
Wins 3 of 7 areas
Reasoning · Images and charts · Following instructions
Claude Sonnet 4 wins more areas, narrowly.EXAONE 4.5 33B is better at reasoning.
Scores updated · 25 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 software31Claude Sonnet 43 of 4 tests
- AgentsCarrying out multi-step tasks on its own20Claude Sonnet 42 of 2 tests
- FactsGetting facts right instead of making them up20Claude Sonnet 42 of 2 tests
- Long documentsFinding answers in very long texts10Claude Sonnet 41 of 1 test
- ReasoningHard problems that need careful thinking02EXAONE 4.5 33B2 of 3 tests · 1 tie
- Images and chartsUnderstanding pictures, charts and video02EXAONE 4.5 33B2 of 2 tests
- Following instructionsDoing exactly what it is asked01EXAONE 4.5 33B1 of 1 test
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.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where Claude Sonnet 4 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+50.8points ahead
- Reasons across sets of long documentsAA-LCR+15.6points ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+15points ahead
Where EXAONE 4.5 33B pulls ahead
- Recent programming contest problemsLiveCodeBench v6+32.9points ahead
- Harder college exam questions with imagesMMMU-Pro+5.5points ahead
- College exam questions with charts, maps and diagramsMMMU+4.3points ahead
Every test, side by side
All 25 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingClaude Sonnet 4
- LiveCodeBench v6EXAONE 4.5 33B by 32.948.581.4+32.9
- Terminal-Bench 2.1Claude Sonnet 4 by 1536.321.3+15
- SciCodeClaude Sonnet 4 by 124028+12
- Terminal-Bench HardClaude Sonnet 4 by 10.631.120.5+10.6
AgentsClaude Sonnet 4
- GDPValClaude Sonnet 4 by 9.79.70+9.7
- τ-Bench V3 · BankingClaude Sonnet 4 by 5.616.711.1+5.6
ReasoningEXAONE 4.5 33B
- Humanity's Last ExamEXAONE 4.5 33B by 2.210.712.9+2.2
- GPQA DiamondEXAONE 4.5 33B by 1.777.779.4+1.7
- CritPttie0.30.3tie
FactsClaude Sonnet 4
- AA-Omniscience · Non-hallucinationClaude Sonnet 4 by 50.870.920.1+50.8
- AA-Omniscience · AccuracyClaude Sonnet 4 by 6.522.716.2+6.5
Images and chartsEXAONE 4.5 33B
Following instructionsEXAONE 4.5 33B
- IFBenchEXAONE 4.5 33B by 3.354.758+3.3
Other results10 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.
- AA-OmniscienceClaude Sonnet 4 by 50.90.2-50.7+50.9
- AIME 2025EXAONE 4.5 33B by 22.470.592.9+22.4
- Artificial Analysis Coding IndexClaude Sonnet 4 by 1437.623.6+14
- τ²-Bench Telecom (AA run)EXAONE 4.5 33B by 13.564.678.1+13.5
- AA Agentic IndexClaude Sonnet 4 by 7.917.69.7+7.9
- AA IntelligenceClaude Sonnet 4 by 5.718.913.2+5.7
- τ²-Bench (Retail)EXAONE 4.5 33B by 2.97577.9+2.9
- IFEvalEXAONE 4.5 33B by 287.689.6+2
- Tau2 airlineEXAONE 4.5 33B by 155.556.5+1
- MMLU-Protie83.783.3tie
Questions people ask
Which is better, Claude Sonnet 4 or EXAONE 4.5 33B?
Claude Sonnet 4 wins four of the seven areas where both have results: coding, agents, facts and long documents. EXAONE 4.5 33B wins reasoning, images and charts and following instructions.
Which is better for coding?
Claude Sonnet 4. It wins 3 of the 4 coding tests both models report; EXAONE 4.5 33B wins 1.
How do you compare the two?
We use the 25 benchmark tests both models have published scores on. The verdict counts the 15 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 10 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.