A.X-K2 is the stronger all-rounder.G9v3-39A5B is better at agents.
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.
- ReasoningHard problems that need careful thinking30A.X-K23 of 3 tests
- CodingWriting and fixing software20A.X-K22 of 2 tests
- Long documentsFinding answers in very long texts10A.X-K21 of 1 test
- AgentsCarrying out multi-step tasks on its own02G9v3-39A5B2 of 2 tests
- 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 A.X-K2 pulls ahead
- Very hard expert questions across many subjectsHumanity's Last Exam+12.1points ahead
- Research-level physics problemsCritPt+8.6points ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+6.4points ahead
Where G9v3-39A5B pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+20points ahead
- Real work tasks from 44 professionsGDPVal+7.3points ahead
- Customer-service tasks in a simulated bankτ-Bench V3 · Banking+6points 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.
CodingA.X-K2
- Terminal-Bench 2.1A.X-K2 by 6.43932.6+6.4
- SciCodeA.X-K2 by 4.24136.8+4.2
AgentsG9v3-39A5B
- GDPValG9v3-39A5B by 7.323.230.5+7.3
- τ-Bench V3 · BankingG9v3-39A5B by 616.122.1+6
ReasoningA.X-K2
- Humanity's Last ExamA.X-K2 by 12.129.617.5+12.1
- CritPtA.X-K2 by 8.68.90.3+8.6
- GPQA DiamondA.X-K2 by 5.285.780.5+5.2
FactsEven
- AA-Omniscience · Non-hallucinationG9v3-39A5B by 206787+20
- AA-Omniscience · AccuracyA.X-K2 by 3.718.614.9+3.7
Other results4 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-OmniscienceG9v3-39A5B by 12-8.23.8+12
- AA Agentic IndexG9v3-39A5B by 7.225.732.9+7.2
- Artificial Analysis Coding IndexA.X-K2 by 5.738.833.1+5.7
- AA Intelligencetie21.521.8tie
Questions people ask
Which is better, A.X-K2 or G9v3-39A5B?
A.X-K2 wins three of the five areas where both have results: coding, reasoning and long documents. G9v3-39A5B wins agents. They are level on facts.
Which is better for coding?
A.X-K2. It wins 2 of the 2 coding tests both models report; G9v3-39A5B wins none.
How do you compare the two?
We use the 14 benchmark tests both models have published scores on. The verdict counts the 10 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 4 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.