GPT-4o vs LFM2.5-1.2B-Instruct
Wins 3 of 6 areas
Coding · Facts · Long documents
Wins 1 of 6 areas
Following instructions
GPT-4o wins more areas, narrowly.LFM2.5-1.2B-Instruct is better at following instructions.
Scores updated · 22 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 software20GPT-4o2 of 2 tests
- FactsGetting facts right instead of making them up20GPT-4o2 of 2 tests
- Long documentsFinding answers in very long texts10GPT-4o1 of 1 test
- Following instructionsDoing exactly what it is asked01LFM2.5-1.2B-Instruct1 of 1 test
- AgentsCarrying out multi-step tasks on its own00Even0 each · 1 tie
- ReasoningHard problems that need careful thinking11Even1 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 tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where GPT-4o pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+47.2points ahead
- Code for real scientific research problemsSciCode+31points ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+20points ahead
Where LFM2.5-1.2B-Instruct pulls ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+7.8points ahead
- Very hard expert questions across many subjectsHumanity's Last Exam+4.9points ahead
Every test, side by side
All 22 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGPT-4o
- SciCodeGPT-4o by 3133.32.3+31
- Terminal-Bench HardGPT-4o by 8.38.30+8.3
ReasoningEven
- GPQA DiamondGPT-4o by 2052.632.6+20
- Humanity's Last ExamLFM2.5-1.2B-Instruct by 4.91.86.7+4.9
- CritPttie00tie
FactsGPT-4o
- AA-Omniscience · Non-hallucinationGPT-4o by 47.262.114.9+47.2
- AA-Omniscience · AccuracyGPT-4o by 16.723.77+16.7
Following instructionsLFM2.5-1.2B-Instruct
- IFBenchLFM2.5-1.2B-Instruct by 7.83643.8+7.8
Other results12 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.
- GSM8KGPT-4o by 86.690.94.3+86.6
- HumanEvalGPT-4o by 84.990.25.3+84.9
- MATH-500 (EM)GPT-4o by 68.874.65.8+68.8
- MATH500 (Pass@1)GPT-4o by 68.674.66+68.6
- AA-OmniscienceGPT-4o by 61.6-10.5-72.1+61.6
- MMLU-ProGPT-4o by 30.374.744.4+30.3
- Avg.LFM2.5-1.2B-Instruct by 22.327.850.1+22.3
- τ²-Bench Telecom (AA run)GPT-4o by 18.128.910.8+18.1
- IFEvalLFM2.5-1.2B-Instruct by 5.28186.2+5.2
- AA Agentic IndexGPT-4o by 4.88.43.6+4.8
- MT-BenchGPT-4o by 4.88.73.9+4.8
- AA IntelligenceGPT-4o by 2.17.35.2+2.1
Questions people ask
Which is better, GPT-4o or LFM2.5-1.2B-Instruct?
GPT-4o wins three of the six areas where both have results: coding, facts and long documents. LFM2.5-1.2B-Instruct wins following instructions. They are level on agents and reasoning.
Which is better for coding?
GPT-4o. It wins 2 of the 2 coding tests both models report; LFM2.5-1.2B-Instruct wins none.
How do you compare the two?
We use the 22 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 12 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.