LIVE INTELLIGENCE · part of the Xentir AI Index

GPT-4o (Aug 2024) vs Qwen2.5-72B (Instruct)

Name: GPT-4o (Aug 2024) vs Qwen2.5-72B (Instruct) — benchmark results
Creator: Epoch AI
License: https://creativecommons.org/licenses/by/4.0/

GPT-4o (Aug 2024) · OpenAI, released 2024-08-06
Qwen2.5-72B (Instruct) OPEN WEIGHTS · Alibaba, released 2024-09-19

“Epoch AI, 'AI Benchmarking Hub'. Published online at epoch.ai. Retrieved from 'https://epoch.ai/benchmarks' [online resource].” · epoch.ai/benchmarks · Licensed CC BY 4.0 · data last verified 2026-09-09 · Methodology · Data Sources

Qwen2.5-72B (Instruct) leads on more of the 4 benchmarks where both are measured — GPT-4o (Aug 2024) on 1, Qwen2.5-72B (Instruct) on 3. The remaining 5 of 9 tracked benchmarks cannot be compared: 3 have a result for only one of the two models, 2 for neither. Nothing on this page is estimated to fill those in.

Head-to-head

Measured on 4 of 9 tracked benchmarks for both models — GPT-4o (Aug 2024) leads on 1, Qwen2.5-72B (Instruct) leads on 3.

Benchmark	GPT-4o (Aug 2024)	Qwen2.5-72B (Instruct)	Leads
Epoch Capabilities Index (ECI)	128.8	129.0	Qwen2.5-72B (Instruct)
SWE-bench Verified	Not measured	Not measured	Not measured on both
GPQA Diamond	0.492	0.491	GPT-4o (Aug 2024)
MATH Level 5	0.533	0.632	Qwen2.5-72B (Instruct)
OTIS Mock AIME 2024-2025	0.064	0.081	Qwen2.5-72B (Instruct)
FrontierMath	0.003	Not measured	Comparison unavailable — Qwen2.5-72B (Instruct) has no published result
FrontierMath Tier 4	Not measured	Not measured	Not measured on both
SimpleQA Verified	0.260	Not measured	Comparison unavailable — Qwen2.5-72B (Instruct) has no published result
Chess Puzzles	0.130	Not measured	Comparison unavailable — Qwen2.5-72B (Instruct) has no published result

Benchmark results measure specific tasks and should not be treated as a universal measure of model quality. A model with no row here for a benchmark has not been measured on it — that is left blank, never estimated.

What remains uncertain

Qwen2.5-72B (Instruct) has no published result on Chess Puzzles, FrontierMath, SimpleQA Verified.
2 tracked benchmarks have no published result for either model.

A missing result means Xentir has not found one published by the source it cites, not that the model performs badly. See Data Sources.

Full profiles

GPT-4o (Aug 2024) → · Qwen2.5-72B (Instruct) → · the full AI Model Rankings table