GPT-4o vs Llama 3.2 Instruct 90B (Vision)
Wins 2 of 3 areas
Coding · Images and charts
Wins 0 of 3 areas
—
GPT-4o is the stronger all-rounder.
Scores updated · 21 tests both models report · How we compare
Where each one wins
Tests won in each of the three 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.
Coding rests on a single test.
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
- Harder college exam questions with imagesMMMU-Pro+16.8points ahead
- College exam questions with charts, maps and diagramsMMMU+10.1points ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+9.4points ahead
Where Llama 3.2 Instruct 90B (Vision) pulls ahead
- Very hard expert questions across many subjectsHumanity's Last Exam+2.7points ahead
Every test, side by side
All 21 tests both models report. The winning score is in its model's colour; marks a score checked independently.
ReasoningEven
- GPQA DiamondGPT-4o by 9.452.643.2+9.4
- Humanity's Last ExamLlama 3.2 Instruct 90B (Vision) by 2.71.84.5+2.7
Images and chartsGPT-4o
Other results14 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.
- MEGA-Bench_macroGPT-4o by 29.549.419.9+29.5
- M-LongDoc (multimodal long-document benchmark)GPT-4o by 27.541.413.9+27.5
- AI2DGPT-4o by 15.394.278.9+15.3
- MATHGPT-4o by 6.285.379.1+6.2
- OlympiadBenchGPT-4o by 5.925.219.3+5.9
- naturalquestions_closedbookGPT-4o by 3.949.645.7+3.9
- MMLUGPT-4o by 3.573.870.3+3.5
- DocVQAGPT-4o by 2.792.890.1+2.7
- GSM8KLlama 3.2 Instruct 90B (Vision) by 2.790.993.6+2.7
- ChartQA_relaxedGPT-4o by 2.688.185.5+2.6
- OpenBookQAGPT-4o by 2.696.894.2+2.6
- NarrativeQAGPT-4o by 1.879.577.7+1.8
- AA Intelligencetie7.36.4tie
- ChartQAtie85.785.5tie
Questions people ask
Which is better, GPT-4o or Llama 3.2 Instruct 90B (Vision)?
GPT-4o wins two of the three areas where both have results: coding and images and charts. Llama 3.2 Instruct 90B (Vision) wins none. They are level on reasoning.
Which is better for coding?
GPT-4o. It wins the one coding test both models report.
How do you compare the two?
We use the 21 benchmark tests both models have published scores on. The verdict counts the 7 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 14 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.