Claude Haiku 4.5 vs K2 Horizon 375B A23B
Wins 0 of 5 areas
—
Wins 5 of 5 areas
Coding · Agents · Reasoning · Facts · Long documents
K2 Horizon 375B A23B is the stronger all-rounder.
Scores updated · 15 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 own03K2 Horizon 375B A23B3 of 3 tests
- ReasoningHard problems that need careful thinking03K2 Horizon 375B A23B3 of 3 tests
- CodingWriting and fixing software01K2 Horizon 375B A23B1 of 2 tests · 1 tie
- FactsGetting facts right instead of making them up01K2 Horizon 375B A23B1 of 2 tests · 1 tie
- Long documentsFinding answers in very long texts01K2 Horizon 375B A23B1 of 1 test
Long documents rests on a single test.
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 K2 Horizon 375B A23B pulls ahead
- Real work tasks from 44 professionsGDPVal+31.4points ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+27.7points ahead
- Customer-service tasks in a simulated bankτ-Bench V3 · Banking+24.9points ahead
Where Claude Haiku 4.5 pulls ahead
No clear win on a test scored out of 100.
Every test, side by side
All 15 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingK2 Horizon 375B A23B
- Terminal-Bench 2.1K2 Horizon 375B A23B by 27.744.271.9+27.7
- SciCodetie42.242.9tie
AgentsK2 Horizon 375B A23B
- GDPValK2 Horizon 375B A23B by 31.411.643+31.4
- τ-Bench V3 · BankingK2 Horizon 375B A23B by 24.99.334.2+24.9
- Terminal-Bench 4.0K2 Horizon 375B A23B by 1.501.5+1.5
ReasoningK2 Horizon 375B A23B
- Humanity's Last ExamK2 Horizon 375B A23B by 21.610.432+21.6
- GPQA DiamondK2 Horizon 375B A23B by 20.167.287.3+20.1
- CritPtK2 Horizon 375B A23B by 4.604.6+4.6
FactsK2 Horizon 375B A23B
- AA-Omniscience · Non-hallucinationK2 Horizon 375B A23B by 1.472.774.1+1.4
- AA-Omniscience · Accuracytie1818.2tie
Long documentsK2 Horizon 375B A23B
- AA-LCRK2 Horizon 375B A23B by 5.774.380+5.7
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 Agentic IndexK2 Horizon 375B A23B by 30.610.340.9+30.6
- Artificial Analysis Coding IndexK2 Horizon 375B A23B by 17.643.961.5+17.6
- AA IntelligenceK2 Horizon 375B A23B by 13.616.930.5+13.6
- AA-OmniscienceK2 Horizon 375B A23B by 1.4-4.4-3+1.4
Questions people ask
Which is better, Claude Haiku 4.5 or K2 Horizon 375B A23B?
K2 Horizon 375B A23B wins all five areas where both have results: coding, agents, reasoning, facts and long documents. Claude Haiku 4.5 wins none.
Which is better for coding?
K2 Horizon 375B A23B. It wins 1 of the 2 coding tests both models report; Claude Haiku 4.5 wins none, and 1 is a tie.
How do you compare the two?
We use the 15 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 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.