Claude 3.5 Haiku vs ERNIE 4.5 300B A47B
Wins 2 of 5 areas
Long documents · Following instructions
Wins 2 of 5 areas
Coding · Reasoning
The two are evenly matched.Claude 3.5 Haiku is better at long documents and following instructions; ERNIE 4.5 300B A47B at coding and reasoning.
Scores updated · 14 tests both models report · How we compare
Where each one wins
Tests won in each of the five 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 software02ERNIE 4.5 300B A47B2 of 2 tests
- ReasoningHard problems that need careful thinking01ERNIE 4.5 300B A47B1 of 3 tests · 2 ties
- Long documentsFinding answers in very long texts10Claude 3.5 Haiku1 of 1 test
- Following instructionsDoing exactly what it is asked10Claude 3.5 Haiku1 of 1 test
- FactsGetting facts right instead of making them up11Even1 each
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 4.5 300B A47B costs 71% less for the same work.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where ERNIE 4.5 300B A47B pulls ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+40.3points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+5.9points ahead
- Code for real scientific research problemsSciCode+4.1points ahead
Where Claude 3.5 Haiku pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+25.8points ahead
- Reasons across sets of long documentsAA-LCR+23.3points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+3.7points ahead
Every test, side by side
All 14 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingERNIE 4.5 300B A47B
- SciCodeERNIE 4.5 300B A47B by 4.127.431.5+4.1
- Terminal-Bench HardERNIE 4.5 300B A47B by 3.82.36.1+3.8
ReasoningERNIE 4.5 300B A47B
- GPQA DiamondERNIE 4.5 300B A47B by 40.340.881.1+40.3
- Humanity's Last Examtie3.63.3tie
- CritPttie00tie
FactsEven
- AA-Omniscience · Non-hallucinationClaude 3.5 Haiku by 25.858.933.1+25.8
- AA-Omniscience · AccuracyERNIE 4.5 300B A47B by 5.913.219.1+5.9
Following instructionsClaude 3.5 Haiku
- IFBenchClaude 3.5 Haiku by 3.742.839.1+3.7
Other results5 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.
- τ²-Bench Telecom (AA run)Claude 3.5 Haiku by 24.624.60+24.6
- AA-OmniscienceClaude 3.5 Haiku by 12.6-22.5-35.1+12.6
- AA Agentic IndexClaude 3.5 Haiku by 1.91.90+1.9
- AA IntelligenceClaude 3.5 Haiku by 1.48.97.5+1.4
- Artificial Analysis Coding IndexClaude 3.5 Haiku by 1.415.914.5+1.4
Questions people ask
Which is better, Claude 3.5 Haiku or ERNIE 4.5 300B A47B?
Claude 3.5 Haiku and ERNIE 4.5 300B A47B each win two of the five areas where both have results. Claude 3.5 Haiku wins long documents and following instructions; ERNIE 4.5 300B A47B wins coding and reasoning. ERNIE 4.5 300B A47B costs 71% less. They are level on facts.
Which is better for coding?
ERNIE 4.5 300B A47B. It wins 2 of the 2 coding tests both models report; Claude 3.5 Haiku wins none.
Which is cheaper?
Claude 3.5 Haiku costs $0.80 per million input tokens and $4.00 per million output tokens; ERNIE 4.5 300B A47B costs $0.28 and $1.10. That makes ERNIE 4.5 300B A47B about 71% cheaper for the same work.
How do you compare the two?
We use the 14 benchmark tests both models have published scores on. The verdict counts the 9 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 5 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.