o4-mini vs Qwen2.5 Instruct 72B
Wins 5 of 5 areas
Coding · Reasoning · Facts · Long documents · Following instructions
Wins 0 of 5 areas
—
o4-mini is the stronger all-rounder.Qwen2.5 Instruct 72B is cheaper.
Scores updated · 21 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.
- CodingWriting and fixing software30o4-mini3 of 3 tests
- ReasoningHard problems that need careful thinking20o4-mini2 of 3 tests · 1 tie
- FactsGetting facts right instead of making them up20o4-mini2 of 2 tests
- Long documentsFinding answers in very long texts10o4-mini1 of 1 test
- Following instructionsDoing exactly what it is asked10o4-mini1 of 1 test
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.
Qwen2.5 Instruct 72B costs 82% 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 o4-mini pulls ahead
- Fixes real GitHub issues in Python projectsSWE-bench Verified+44.3points ahead
- Reasons across sets of long documentsAA-LCR+41points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+31.8points ahead
Where Qwen2.5 Instruct 72B pulls ahead
No clear win on a test scored out of 100.
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.
Codingo4-mini
- SWE-bench Verifiedo4-mini by 44.368.123.8+44.3
- SciCodeo4-mini by 19.846.526.7+19.8
- Terminal-Bench Hardo4-mini by 10.715.24.5+10.7
Reasoningo4-mini
- GPQA Diamondo4-mini by 29.378.449.1+29.3
- Humanity's Last Examo4-mini by 12.916.53.6+12.9
- CritPttie0.60tie
Factso4-mini
- AA-Omniscience · Accuracyo4-mini by 7.324.817.5+7.3
- AA-Omniscience · Non-hallucinationo4-mini by 519.514.5+5
Other results11 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.
- HarmBencho4-mini by 24.29772.8+24.2
- LiveCodeBencho4-mini by 21.977.455.5+21.9
- τ²-Bench Telecom (AA run)o4-mini by 21.155.634.5+21.1
- AIR-Bench 2024o4-mini by 19.578.559+19.5
- AA-Omniscienceo4-mini by 17.4-35.7-53.1+17.4
- Artificial Analysis Coding Indexo4-mini by 13.725.611.9+13.7
- AA Intelligenceo4-mini by 916.77.7+9
- anthropic_red_teamQwen2.5 Instruct 72B by 1.498.299.6+1.4
- bbqQwen2.5 Instruct 72B by 1.49495.4+1.4
- XSTesttie97.497.9tie
- simple_safety_teststie100100tie
Questions people ask
Which is better, o4-mini or Qwen2.5 Instruct 72B?
o4-mini wins all five areas where both have results: coding, reasoning, facts, long documents and following instructions. Qwen2.5 Instruct 72B wins none, but costs 82% less.
Which is better for coding?
o4-mini. It wins 3 of the 3 coding tests both models report; Qwen2.5 Instruct 72B wins none.
Which is cheaper?
o4-mini costs $1.10 per million input tokens and $4.40 per million output tokens; Qwen2.5 Instruct 72B costs $0.47 and $0.49. That makes Qwen2.5 Instruct 72B about 82% cheaper for the same work.
How do you compare the two?
We use the 21 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 11 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.