Claude 3 Haiku vs LFM2-24B-A2B
Wins 2 of 6 areas
Coding · Long documents
Wins 2 of 6 areas
Reasoning · Following instructions
The two are evenly matched.Claude 3 Haiku is better at coding and long documents; LFM2-24B-A2B at reasoning and following instructions.
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.
- CodingWriting and fixing software10Claude 3 Haiku1 of 2 tests · 1 tie
- Long documentsFinding answers in very long texts10Claude 3 Haiku1 of 1 test
- ReasoningHard problems that need careful thinking01LFM2-24B-A2B1 of 3 tests · 2 ties
- Following instructionsDoing exactly what it is asked01LFM2-24B-A2B1 of 1 test
- AgentsCarrying out multi-step tasks on its own00Even0 each · 1 tie
- FactsGetting facts right instead of making them up11Even1 each
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.
LFM2-24B-A2B costs 90% less for the same work.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where Claude 3 Haiku pulls ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+11.1points ahead
- Code for real scientific research problemsSciCode+7.7points ahead
Where LFM2-24B-A2B pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+11.5points ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+10points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+9.8points 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.
CodingClaude 3 Haiku
- SciCodeClaude 3 Haiku by 7.718.610.9+7.7
- Terminal-Bench Hardtie0.80tie
ReasoningLFM2-24B-A2B
- GPQA DiamondLFM2-24B-A2B by 1037.447.4+10
- Humanity's Last Examtie4.14.2tie
- CritPttie00tie
FactsEven
- AA-Omniscience · Non-hallucinationLFM2-24B-A2B by 11.519.531+11.5
- AA-Omniscience · AccuracyClaude 3 Haiku by 11.117.66.5+11.1
Following instructionsLFM2-24B-A2B
- IFBenchLFM2-24B-A2B by 9.836.145.9+9.8
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.
- τ²-Bench Telecom (AA run)Claude 3 Haiku by 1021.111.1+10
- AA-OmniscienceClaude 3 Haiku by 9.5-48.6-58.1+9.5
- AA Agentic IndexClaude 3 Haiku by 3.373.7+3.3
- Artificial Analysis Coding IndexClaude 3 Haiku by 3.16.73.6+3.1
- AA Intelligencetie5.65.9tie
Questions people ask
Which is better, Claude 3 Haiku or LFM2-24B-A2B?
Claude 3 Haiku and LFM2-24B-A2B each win two of the six areas where both have results. Claude 3 Haiku wins coding and long documents; LFM2-24B-A2B wins reasoning and following instructions. LFM2-24B-A2B costs 90% less. They are level on agents and facts.
Which is better for coding?
Claude 3 Haiku. It wins 1 of the 2 coding tests both models report; LFM2-24B-A2B wins none, and 1 is a tie.
Which is cheaper?
Claude 3 Haiku costs $0.25 per million input tokens and $1.25 per million output tokens; LFM2-24B-A2B costs $0.03 and $0.12. That makes LFM2-24B-A2B about 90% 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 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.