GPT-4.1: charts and benchmarks
Every chart we tracked that shows GPT-4.1, from the lab itself and from the trackers people trust, with how far each one travelled.
GPT-4.1 on its own benchmark table
Drawn from the numbers openrouter.ai published as text (Choosing the Optimal Image Input Detail Level in LLMs, read Jul 7, 2026). Each panel is one test; the highlighted bar is GPT-4.1. Source: openrouter.ai.
See the numbers
| Accuracy | gemini-3.5-flash 80.1, gpt-5.5 79, gemini-3.1-pro 78.4, gemini-3.5-flash 77.9, gemini-3.1-pro 75.5, gpt-5.5 65.2, gpt-4.1 57.5, gpt-5.4-mini 55.8, gpt-5.4-mini 46.1, gpt-4.1 40.1 |
|---|---|
| Cost / question | gemini-3.1-pro 11.12, gemini-3.1-pro 9.53, gpt-5.5 5.1, gpt-5.5 4.5, gemini-3.5-flash 2.96, gemini-3.5-flash 2.8, gpt-4.1 0.66, gpt-4.1 0.43, gpt-5.4-mini 0.14, gpt-5.4-mini 0.08 |
| Reasoning tok/req | gemini-3.1-pro 6964, gemini-3.1-pro 6344, gemini-3.5-flash 2876, gemini-3.5-flash 2602, gpt-5.5 1180, gpt-5.5 730, gpt-5.4-mini 0, gpt-5.4-mini 0, gpt-4.1 0, gpt-4.1 0 |
Official charts from OpenAI
As published on X, Hugging Face, GitHub and their blog. Each card opens our own page for the chart.
More charts that feature GPT-4.1
Benchmark trackers, other labs and researchers. Each card opens our own page for the chart.
Make it yours
Your numbers, in a look-alike of the original.
Starts from the real numbers of "Accuracy / Percentile (%)" by deepseek. Replace every number with yours before you post. Make one like this, step by step
Look-alike score 8.3 of 10. Free pictures carry a small "Made with ViralChart" line. That line is how other people find us.
Questions people ask
What charts show GPT-4.1?
We hold 20 charts that show GPT-4.1, 4 of them published by OpenAI. The newest is "Terminal-Bench AI Model Scores" (Sep 28, 2026). Each card opens our own page for the chart.
Who publishes GPT-4.1 benchmark charts?
OpenAI, @ArtificialAnlys, @EpochAIResearch, @AnthropicAI, @Alibaba_Qwen, openrouter, moonshotai/Kimi-K2-Instruct