Claude 3 Haiku vs Falcon H1R 7B
Wins 2 of 6 areas
Facts · Long documents
Wins 3 of 6 areas
Coding · Reasoning · Following instructions
Falcon H1R 7B wins more areas, narrowly.Claude 3 Haiku is better at facts.
Scores updated · 15 tests both models report · How we compare
Where each one wins
Tests won in each of the six 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 thinking02Falcon H1R 7B2 of 3 tests · 1 tie
- CodingWriting and fixing software02Falcon H1R 7B2 of 2 tests
- Following instructionsDoing exactly what it is asked01Falcon H1R 7B1 of 1 test
- FactsGetting facts right instead of making them up20Claude 3 Haiku2 of 2 tests
- Long documentsFinding answers in very long texts10Claude 3 Haiku1 of 1 test
- AgentsCarrying out multi-step tasks on its own00Even0 each · 1 tie
Agents, long documents and following instructions rest on a single test each.
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 Falcon H1R 7B pulls ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+28.7points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+18.3points ahead
- Very hard expert questions across many subjectsHumanity's Last Exam+6.9points ahead
Where Claude 3 Haiku pulls ahead
- Reasons across sets of long documentsAA-LCR+16.7points ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+8.3points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+3points 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.
CodingFalcon H1R 7B
- SciCodeFalcon H1R 7B by 6.318.624.9+6.3
- Terminal-Bench HardFalcon H1R 7B by 1.50.82.3+1.5
ReasoningFalcon H1R 7B
- GPQA DiamondFalcon H1R 7B by 28.737.466.1+28.7
- Humanity's Last ExamFalcon H1R 7B by 6.94.111+6.9
- CritPttie00.3tie
FactsClaude 3 Haiku
- AA-Omniscience · Non-hallucinationClaude 3 Haiku by 8.319.511.2+8.3
- AA-Omniscience · AccuracyClaude 3 Haiku by 317.614.6+3
Following instructionsFalcon H1R 7B
- IFBenchFalcon H1R 7B by 18.336.154.4+18.3
Other results5 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 3 Haiku by 12.7-48.6-61.3+12.7
- τ²-Bench Telecom (AA run)Falcon H1R 7B by 6.721.127.8+6.7
- Artificial Analysis Coding IndexFalcon H1R 7B by 3.16.79.8+3.1
- AA Agentic IndexFalcon H1R 7B by 2.379.3+2.3
- AA IntelligenceFalcon H1R 7B by 2.25.67.8+2.2
Questions people ask
Which is better, Claude 3 Haiku or Falcon H1R 7B?
Falcon H1R 7B wins three of the six areas where both have results: coding, reasoning and following instructions. Claude 3 Haiku wins facts and long documents. They are level on agents.
Which is better for coding?
Falcon H1R 7B. It wins 2 of the 2 coding tests both models report; Claude 3 Haiku wins none.
How do you compare the two?
We use the 15 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 5 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.