K2 Horizon 36B-A4B wins more areas, narrowly.A.X-K2 is better at reasoning.
Scores updated · 13 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.
- AgentsCarrying out multi-step tasks on its own02K2 Horizon 36B-A4B2 of 3 tests · 1 tie
- Long documentsFinding answers in very long texts01K2 Horizon 36B-A4B1 of 1 test
- ReasoningHard problems that need careful thinking30A.X-K23 of 3 tests
- CodingWriting and fixing software11Even1 each
- FactsGetting facts right instead of making them up11Even1 each
Long documents rests on a single test.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where K2 Horizon 36B-A4B pulls ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+17.6points ahead
- Customer-service tasks in a simulated bankτ-Bench V3 · Banking+15.9points ahead
- Real work tasks from 44 professionsGDPVal+11.5points ahead
Where A.X-K2 pulls ahead
- Research-level physics problemsCritPt+6.6points ahead
- Very hard expert questions across many subjectsHumanity's Last Exam+6.2points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+3.6points ahead
Every test, side by side
All 13 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingEven
- Terminal-Bench 2.1K2 Horizon 36B-A4B by 17.63956.6+17.6
- SciCodeA.X-K2 by 14140+1
AgentsK2 Horizon 36B-A4B
- τ-Bench V3 · BankingK2 Horizon 36B-A4B by 15.916.132+15.9
- GDPValK2 Horizon 36B-A4B by 11.523.234.7+11.5
- Terminal-Bench 4.0tie00tie
ReasoningA.X-K2
- CritPtA.X-K2 by 6.68.92.3+6.6
- Humanity's Last ExamA.X-K2 by 6.229.623.4+6.2
- GPQA DiamondA.X-K2 by 3.585.782.2+3.5
FactsEven
- AA-Omniscience · AccuracyA.X-K2 by 3.618.615+3.6
- AA-Omniscience · Non-hallucinationK2 Horizon 36B-A4B by 2.56769.5+2.5
Long documentsK2 Horizon 36B-A4B
- AA-LCRK2 Horizon 36B-A4B by 1.77071.7+1.7
Other results2 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 IntelligenceK2 Horizon 36B-A4B by 3.821.525.3+3.8
- AA-OmniscienceA.X-K2 by 2.7-8.2-10.9+2.7
Questions people ask
Which is better, A.X-K2 or K2 Horizon 36B-A4B?
K2 Horizon 36B-A4B wins two of the five areas where both have results: agents and long documents. A.X-K2 wins reasoning. They are level on coding and facts.
Which is better for coding?
Neither. They win 1 coding test each of the 2 both models report.
How do you compare the two?
We use the 13 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 2 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.