The Model Index
Up to four models

Qwen2.5 vs GPT-4o

On the index (GPT-4 = 100), Qwen2.5 scores 112 and GPT-4o 111. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.

Qwen2.5GPT-4o
Next to Qwen2.5
Presets
Qwen2.5
11296–128
WeightsOpen
ReleasedSep 2024
Price in / out$0.36 / $0.40
Context33K
Core tests taken2 of 9
GPT-4o
11195–127
WeightsClosed
ReleasedMay 2024
Price in / out$2.5 / $10
Context128K
Core tests taken2 of 9

Qwen2.5 and GPT-4o are too close to call (1 point apart, within the uncertainty), ahead on 1 of 2 shared core tests, and costs 12× less.

Test by testthird-party results

Qwen2.5GPT-4o
0%25%50%75%100%REASONING & KNOWLEDGEGPQA Diamond (Epoch AI)GPQA Diamond (Epoch AI)MMLU (HELM runs)MMLU (HELM runs)MATHSAIME-style maths (OTIS mock)AIME-style maths (OTIS mock)MATH (level 5)MATH (level 5)AGENTSBALROGBALROGDTBenchDTBenchLMCALMCA
Every result as a table
TestQwen2.5GPT-4o
Reasoning & knowledge
GPQA Diamond (Epoch AI)49.1%48.9%
MMLU (HELM runs)83.4%84.2%
Maths
AIME-style maths (OTIS mock)8.1%6.3%
MATH (level 5)63.2%51.0%
Agents
BALROG16.2%32.3%
DTBench62.9%64.5%
LMCA13.4%16.6%
Task length · METR21 sec24 sec

Tests at least two of these models have taken. Bold rows are the core tests.

This page as a chartDownload PNG