Claude Haiku 4.5 vs K2 Horizon 3.7B
Wins 3 of 5 areas
Coding · Facts · Long documents
Wins 2 of 5 areas
Agents · Reasoning
Claude Haiku 4.5 is the stronger all-rounder.K2 Horizon 3.7B is better at agents.
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.
- CodingWriting and fixing software20Claude Haiku 4.52 of 2 tests
- FactsGetting facts right instead of making them up20Claude Haiku 4.52 of 2 tests
- Long documentsFinding answers in very long texts10Claude Haiku 4.51 of 1 test
- AgentsCarrying out multi-step tasks on its own02K2 Horizon 3.7B2 of 3 tests · 1 tie
- ReasoningHard problems that need careful thinking02K2 Horizon 3.7B2 of 3 tests · 1 tie
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 three tests each model wins by the widest margin. Scores are out of 100.
Where Claude Haiku 4.5 pulls ahead
- Code for real scientific research problemsSciCode+20.2points ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+16.8points ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+16.1points ahead
Where K2 Horizon 3.7B pulls ahead
- Customer-service tasks in a simulated bankτ-Bench V3 · Banking+10.3points ahead
- Real work tasks from 44 professionsGDPVal+7.4points ahead
- Very hard expert questions across many subjectsHumanity's Last Exam+3.5points 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.
CodingClaude Haiku 4.5
- SciCodeClaude Haiku 4.5 by 20.242.222+20.2
- Terminal-Bench 2.1Claude Haiku 4.5 by 16.144.228.1+16.1
AgentsK2 Horizon 3.7B
- τ-Bench V3 · BankingK2 Horizon 3.7B by 10.39.319.6+10.3
- GDPValK2 Horizon 3.7B by 7.411.819.2+7.4
- Terminal-Bench 4.0tie00tie
ReasoningK2 Horizon 3.7B
- Humanity's Last ExamK2 Horizon 3.7B by 3.510.413.9+3.5
- GPQA DiamondK2 Horizon 3.7B by 267.269.2+2
- CritPttie00tie
FactsClaude Haiku 4.5
- AA-Omniscience · Non-hallucinationClaude Haiku 4.5 by 16.872.755.9+16.8
- AA-Omniscience · AccuracyClaude Haiku 4.5 by 7.21810.8+7.2
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-OmniscienceClaude Haiku 4.5 by 24.2-4.4-28.6+24.2
- AA IntelligenceClaude Haiku 4.5 by 1.316.915.6+1.3
Questions people ask
Which is better, Claude Haiku 4.5 or K2 Horizon 3.7B?
Claude Haiku 4.5 wins three of the five areas where both have results: coding, facts and long documents. K2 Horizon 3.7B wins agents and reasoning.
Which is better for coding?
Claude Haiku 4.5. It wins 2 of the 2 coding tests both models report; K2 Horizon 3.7B wins none.
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.