Apriel-v1.5-15B-Thinker vs K2-V2 (high)
Wins 3 of 5 areas
Coding · Reasoning · Following instructions
Wins 1 of 5 areas
Long documents
Apriel-v1.5-15B-Thinker is the stronger all-rounder.K2-V2 (high) is better at long documents.
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 thinking30Apriel-v1.5-15B-Thinker3 of 3 tests
- CodingWriting and fixing software10Apriel-v1.5-15B-Thinker1 of 2 tests · 1 tie
- Following instructionsDoing exactly what it is asked10Apriel-v1.5-15B-Thinker1 of 1 test
- Long documentsFinding answers in very long texts01K2-V2 (high)1 of 1 test
- FactsGetting facts right instead of making them up11Even1 each
Long documents and following instructions rest on a single test each.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where Apriel-v1.5-15B-Thinker pulls ahead
- Code for real scientific research problemsSciCode+6.2points ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+5.7points ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+3.2points ahead
Where K2-V2 (high) pulls ahead
- Reasons across sets of long documentsAA-LCR+13.7points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+2points 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-v1.5-15B-Thinker
- SciCodeApriel-v1.5-15B-Thinker by 6.234.828.6+6.2
- Terminal-Bench Hardtie10.69.8tie
ReasoningApriel-v1.5-15B-Thinker
- GPQA DiamondApriel-v1.5-15B-Thinker by 3.271.368.1+3.2
- Humanity's Last ExamApriel-v1.5-15B-Thinker by 1.612.110.5+1.6
- CritPtApriel-v1.5-15B-Thinker by 1.11.10+1.1
FactsEven
- AA-Omniscience · Non-hallucinationApriel-v1.5-15B-Thinker by 5.714.79+5.7
- AA-Omniscience · AccuracyK2-V2 (high) by 216.418.5+2
Following instructionsApriel-v1.5-15B-Thinker
- IFBenchApriel-v1.5-15B-Thinker by 1.661.760.1+1.6
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-v1.5-15B-Thinker by 40.668.427.8+40.6
- AA IntelligenceApriel-v1.5-15B-Thinker by 3.913.89.9+3.9
- AA-Omnisciencetie-54.8-55.6tie
Questions people ask
Which is better, Apriel-v1.5-15B-Thinker or K2-V2 (high)?
Apriel-v1.5-15B-Thinker wins three of the five areas where both have results: coding, reasoning and following instructions. K2-V2 (high) wins long documents. They are level on facts.
Which is better for coding?
Apriel-v1.5-15B-Thinker. It wins 1 of the 2 coding tests both models report; K2-V2 (high) wins none, and 1 is a tie.
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.