Claude Haiku 4.5 vs K2 Horizon 36B-A4B
Wins 2 of 5 areas
Facts · Long documents
Wins 2 of 5 areas
Agents · Reasoning
The two are evenly matched.Claude Haiku 4.5 is better at facts and long documents; K2 Horizon 36B-A4B at reasoning and 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.
- ReasoningHard problems that need careful thinking03K2 Horizon 36B-A4B3 of 3 tests
- AgentsCarrying out multi-step tasks on its own02K2 Horizon 36B-A4B2 of 3 tests · 1 tie
- 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
- CodingWriting and fixing software11Even1 each
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 K2 Horizon 36B-A4B pulls ahead
- Real work tasks from 44 professionsGDPVal+22.9points ahead
- Customer-service tasks in a simulated bankτ-Bench V3 · Banking+22.7points ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+15points ahead
Where Claude Haiku 4.5 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+3.2points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+3points ahead
- Reasons across sets of long documentsAA-LCR+2.6points 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 12.444.256.6+12.4
- SciCodeClaude Haiku 4.5 by 2.242.240+2.2
AgentsK2 Horizon 36B-A4B
- GDPValK2 Horizon 36B-A4B by 22.911.834.7+22.9
- τ-Bench V3 · BankingK2 Horizon 36B-A4B by 22.79.332+22.7
- Terminal-Bench 4.0tie00tie
ReasoningK2 Horizon 36B-A4B
- GPQA DiamondK2 Horizon 36B-A4B by 1567.282.2+15
- Humanity's Last ExamK2 Horizon 36B-A4B by 1310.423.4+13
- CritPtK2 Horizon 36B-A4B by 2.302.3+2.3
FactsClaude Haiku 4.5
- AA-Omniscience · Non-hallucinationClaude Haiku 4.5 by 3.272.769.5+3.2
- AA-Omniscience · AccuracyClaude Haiku 4.5 by 31815+3
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 8.416.925.3+8.4
- AA-OmniscienceClaude Haiku 4.5 by 6.5-4.4-10.9+6.5
Questions people ask
Which is better, Claude Haiku 4.5 or K2 Horizon 36B-A4B?
Claude Haiku 4.5 and K2 Horizon 36B-A4B each win two of the five areas where both have results. Claude Haiku 4.5 wins facts and long documents; K2 Horizon 36B-A4B wins agents and reasoning. They are level on coding.
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.