Claude 3.5 Haiku vs G9v3-3B
Wins 1 of 5 areas
Coding
Wins 3 of 5 areas
Agents · Reasoning · Long documents
G9v3-3B is the stronger all-rounder.Claude 3.5 Haiku is better at coding.
Scores updated · 14 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 thinking01G9v3-3B1 of 3 tests · 2 ties
- AgentsCarrying out multi-step tasks on its own01G9v3-3B1 of 2 tests · 1 tie
- Long documentsFinding answers in very long texts01G9v3-3B1 of 1 test
- CodingWriting and fixing software20Claude 3.5 Haiku2 of 2 tests
- FactsGetting facts right instead of making them up11Even1 each
Long documents rests on a single test.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where G9v3-3B pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+29.4points ahead
- Reasons across sets of long documentsAA-LCR+14points ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+3points ahead
Where Claude 3.5 Haiku pulls ahead
- Code for real scientific research problemsSciCode+9.7points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+6.7points ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+4.1points ahead
Every test, side by side
All 14 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingClaude 3.5 Haiku
- SciCodeClaude 3.5 Haiku by 9.727.417.7+9.7
- Terminal-Bench 2.1Claude 3.5 Haiku by 4.110.16+4.1
ReasoningG9v3-3B
- GPQA DiamondG9v3-3B by 340.843.8+3
- Humanity's Last Examtie3.64.5tie
- CritPttie00tie
FactsEven
- AA-Omniscience · Non-hallucinationG9v3-3B by 29.458.988.3+29.4
- AA-Omniscience · AccuracyClaude 3.5 Haiku by 6.713.26.5+6.7
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-OmniscienceG9v3-3B by 18.1-22.5-4.4+18.1
- AA Agentic IndexG9v3-3B by 12.11.914+12.1
- Artificial Analysis Coding IndexClaude 3.5 Haiku by 615.99.9+6
- AA IntelligenceG9v3-3B by 1.98.910.8+1.9
Questions people ask
Which is better, Claude 3.5 Haiku or G9v3-3B?
G9v3-3B wins three of the five areas where both have results: agents, reasoning and long documents. Claude 3.5 Haiku wins coding. They are level on facts.
Which is better for coding?
Claude 3.5 Haiku. It wins 2 of the 2 coding tests both models report; G9v3-3B wins none.
How do you compare the two?
We use the 14 benchmark tests both models have published scores on. The verdict counts the 10 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.