LongCat-2.0 wins more areas, narrowly.Claude Haiku 4.5 is better at long documents.
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.
- ReasoningHard problems that need careful thinking03LongCat-2.03 of 3 tests
- AgentsCarrying out multi-step tasks on its own02LongCat-2.02 of 3 tests · 1 tie
- Long documentsFinding answers in very long texts10Claude Haiku 4.51 of 1 test
- CodingWriting and fixing software11Even1 each
- FactsGetting facts right instead of making them up11Even1 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.
LongCat-2.0 costs 75% less for the same work.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where LongCat-2.0 pulls ahead
- Very hard expert questions across many subjectsHumanity's Last Exam+23.3points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+11.6points ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+10.8points ahead
Where Claude Haiku 4.5 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+48.1points ahead
- Reasons across sets of long documentsAA-LCR+9.3points ahead
- Code for real scientific research problemsSciCode+5.9points ahead
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.
CodingEven
- Terminal-Bench 2.1LongCat-2.0 by 644.250.2+6
- SciCodeClaude Haiku 4.5 by 5.942.236.3+5.9
AgentsLongCat-2.0
- GDPValLongCat-2.0 by 6.811.818.6+6.8
- τ-Bench V3 · BankingLongCat-2.0 by 3.99.313.2+3.9
- Terminal-Bench 4.0tie00tie
ReasoningLongCat-2.0
- Humanity's Last ExamLongCat-2.0 by 23.310.433.7+23.3
- GPQA DiamondLongCat-2.0 by 10.867.278+10.8
- CritPtLongCat-2.0 by 2.602.6+2.6
FactsEven
- AA-Omniscience · Non-hallucinationClaude Haiku 4.5 by 48.172.724.6+48.1
- AA-Omniscience · AccuracyLongCat-2.0 by 11.61829.6+11.6
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-OmniscienceClaude Haiku 4.5 by 19-4.4-23.4+19
- AA Agentic IndexLongCat-2.0 by 5.610.315.9+5.6
- AA IntelligenceLongCat-2.0 by 2.216.919.1+2.2
- Artificial Analysis Coding IndexLongCat-2.0 by 1.443.945.3+1.4
Questions people ask
Which is better, Claude Haiku 4.5 or LongCat-2.0?
LongCat-2.0 wins two of the five areas where both have results: agents and reasoning. Claude Haiku 4.5 wins long documents. 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.
Which is cheaper?
Claude Haiku 4.5 costs $1.00 per million input tokens and $5.00 per million output tokens; LongCat-2.0 costs $0.30 and $1.20. That makes LongCat-2.0 about 75% cheaper for the same work.
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.