Claude 3.5 Haiku vs Tri-21B-Think
Wins 3 of 5 areas
Coding · Facts · Long documents
Wins 2 of 5 areas
Reasoning · Following instructions
Claude 3.5 Haiku is the stronger all-rounder.Tri-21B-Think is better at reasoning.
Scores updated · 11 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.
- FactsGetting facts right instead of making them up10Claude 3.5 Haiku1 of 2 tests · 1 tie
- CodingWriting and fixing software10Claude 3.5 Haiku1 of 1 test
- Long documentsFinding answers in very long texts10Claude 3.5 Haiku1 of 1 test
- ReasoningHard problems that need careful thinking02Tri-21B-Think2 of 3 tests · 1 tie
- Following instructionsDoing exactly what it is asked01Tri-21B-Think1 of 1 test
Coding, long documents and following instructions rest on a single test each.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where Claude 3.5 Haiku pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+44.3points ahead
- Reasons across sets of long documentsAA-LCR+11.6points ahead
- Hard command-line tasks in a real terminalTerminal-Bench Hard+1.5points ahead
Where Tri-21B-Think pulls ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+19.3points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+11.8points ahead
- Very hard expert questions across many subjectsHumanity's Last Exam+2.3points ahead
Every test, side by side
All 11 tests both models report. The winning score is in its model's colour; marks a score checked independently.
ReasoningTri-21B-Think
- GPQA DiamondTri-21B-Think by 19.340.860.1+19.3
- Humanity's Last ExamTri-21B-Think by 2.33.65.9+2.3
- CritPttie00.3tie
FactsClaude 3.5 Haiku
- AA-Omniscience · Non-hallucinationClaude 3.5 Haiku by 44.358.914.6+44.3
- AA-Omniscience · Accuracytie13.212.3tie
Long documentsClaude 3.5 Haiku
- AA-LCRClaude 3.5 Haiku by 11.627.315.7+11.6
Following instructionsTri-21B-Think
- IFBenchTri-21B-Think by 11.842.854.6+11.8
Other results3 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.
- τ²-Bench Telecom (AA run)Tri-21B-Think by 56.424.681+56.4
- AA-OmniscienceClaude 3.5 Haiku by 40-22.5-62.5+40
- AA Intelligencetie8.99tie
Questions people ask
Which is better, Claude 3.5 Haiku or Tri-21B-Think?
Claude 3.5 Haiku wins three of the five areas where both have results: coding, facts and long documents. Tri-21B-Think wins reasoning and following instructions.
Which is better for coding?
Claude 3.5 Haiku. It wins the one coding test both models report.
How do you compare the two?
We use the 11 benchmark tests both models have published scores on. The verdict counts the 8 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 3 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.