Apriel-v1.6-15B-Thinker vs Claude 3 Haiku
Wins 4 of 5 areas
Coding · Reasoning · Long documents · Following instructions
Wins 1 of 5 areas
Facts
Apriel-v1.6-15B-Thinker is the stronger all-rounder.Claude 3 Haiku is better at facts.
Scores updated · 12 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.
- ReasoningHard problems that need careful thinking20Apriel-v1.6-15B-Thinker2 of 3 tests · 1 tie
- CodingWriting and fixing software20Apriel-v1.6-15B-Thinker2 of 2 tests
- Long documentsFinding answers in very long texts10Apriel-v1.6-15B-Thinker1 of 1 test
- Following instructionsDoing exactly what it is asked10Apriel-v1.6-15B-Thinker1 of 1 test
- FactsGetting facts right instead of making them up01Claude 3 Haiku1 of 2 tests · 1 tie
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 tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where Apriel-v1.6-15B-Thinker pulls ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+35.9points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+33points ahead
- Reasons across sets of long documentsAA-LCR+22.6points ahead
Where Claude 3 Haiku pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+11.4points ahead
Every test, side by side
All 12 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingApriel-v1.6-15B-Thinker
- SciCodeApriel-v1.6-15B-Thinker by 18.737.318.6+18.7
- Terminal-Bench HardApriel-v1.6-15B-Thinker by 13.614.40.8+13.6
ReasoningApriel-v1.6-15B-Thinker
- GPQA DiamondApriel-v1.6-15B-Thinker by 35.973.337.4+35.9
- Humanity's Last ExamApriel-v1.6-15B-Thinker by 6.710.84.1+6.7
- CritPttie0.30tie
FactsClaude 3 Haiku
- AA-Omniscience · Non-hallucinationClaude 3 Haiku by 11.48.119.5+11.4
- AA-Omniscience · Accuracytie1717.6tie
Long documentsApriel-v1.6-15B-Thinker
- AA-LCRApriel-v1.6-15B-Thinker by 22.650.327.7+22.6
Following instructionsApriel-v1.6-15B-Thinker
- IFBenchApriel-v1.6-15B-Thinker by 3369.136.1+33
Other results3 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)Apriel-v1.6-15B-Thinker by 48.269.321.1+48.2
- AA-OmniscienceClaude 3 Haiku by 10.8-59.4-48.6+10.8
- AA IntelligenceApriel-v1.6-15B-Thinker by 7.813.45.6+7.8
Questions people ask
Which is better, Apriel-v1.6-15B-Thinker or Claude 3 Haiku?
Apriel-v1.6-15B-Thinker wins four of the five areas where both have results: coding, reasoning, long documents and following instructions. Claude 3 Haiku wins facts.
Which is better for coding?
Apriel-v1.6-15B-Thinker. It wins 2 of the 2 coding tests both models report; Claude 3 Haiku wins none.
How do you compare the two?
We use the 12 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 3 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.