memesearch
SWE-Bench Pro Accuracy Chart

SWE-Bench Pro Accuracy Chart

Download Find similar pictures
Text on the meme
SWE-Bench Pro (Public) 60% GPT-5.3-Codex GPT-5.2-Codex GPT-5.2 50% 40% 30% Accuracy Output tokens 0 20,000 40,000 60,000 80,000 100,000
What it means

A dark benchmark chart compares the accuracy of GPT-5.3-Codex, GPT-5.2-Codex, and GPT-5.2 as output-token counts increase. The joke is that the accompanying caption negotiates for a future model with a huge context window and much better retrieval, while the chart highlights current coding-model performance.

Objects
benchmark chart / график бенчмаркаline graph / линейный графикdata points / точки данныхx-axis / ось Xy-axis / ось Yblack background / чёрный фонwhite labels / белые подписиblue lines / синие линииaccuracy scale / шкала точностиoutput token scale / шкала выходных токеновOpenAI logo / логотип OpenAI
Tags
#ai#искусственный интеллект#benchmark#бенчмарк#coding#программирование#gpt#codex#swe-bench#accuracy#точность#context window#контекст#retrieval#токены#chart#график#reaction#реакция#tech meme#технический мем
Source
Type
image
Size
800×589 · 26.1 KiB
Mood
sarcastic, analytical
Language
en
Added

Similar memes

GPT-5.3-Codex Benchmark Flex GPT-5.3-Codex Benchmark Flex o3 Crushes Coding Benchmarks o3 Crushes Coding Benchmarks TOON Benchmark Accuracy Meme TOON Benchmark Accuracy Meme Misleading AI Accuracy Graph Misleading AI Accuracy Graph AI Benchmark Comparison Table AI Benchmark Comparison Table Opus 4.1 SWE-bench Accuracy Opus 4.1 SWE-bench Accuracy o3 Beats Humans at Math and Science o3 Beats Humans at Math and Science BullshitBench Model Improvement Chart BullshitBench Model Improvement Chart AI Model Name Escalation Chart AI Model Name Escalation Chart Opus 5 Benchmark Sweep Opus 5 Benchmark Sweep Opus Dominates Long-Context Retrieval Opus Dominates Long-Context Retrieval GPT-5 Numbers Go Sky-High GPT-5 Numbers Go Sky-High GPT Tool Use Benchmark Chart GPT Tool Use Benchmark Chart AI Felony Bench Scorecard AI Felony Bench Scorecard GPT-6.1 Crushes Opus 5.5 GPT-6.1 Crushes Opus 5.5 More Compute, Mixed Results More Compute, Mixed Results Apple Chips Leave Intel Behind Apple Chips Leave Intel Behind o3 Research Math Accuracy Chart o3 Research Math Accuracy Chart GPT-4.5 Pricing Table Shock GPT-4.5 Pricing Table Shock GPU Usage Spikes to 500 Million GPU Usage Spikes to 500 Million Opus 4.6 Benchmark Sweep Opus 4.6 Benchmark Sweep Active Popes Suddenly Drop Active Popes Suddenly Drop Распределение IQ и тесты кода Распределение IQ и тесты кода Offensive Cyber Evaluations Chart Offensive Cyber Evaluations Chart Llama 4 Behemoth Benchmark Victory Llama 4 Behemoth Benchmark Victory ChatGPT Explains a Neural Network VIDEO 0:29 ChatGPT Explains a Neural Network Humanity's Last Exam Results Humanity's Last Exam Results GPT Levels Past the Engineering Ladder GPT Levels Past the Engineering Ladder DDR5 RAM Price Explosion DDR5 RAM Price Explosion From AM to Iridium Flatline VIDEO 0:30 From AM to Iridium Flatline GitHub Average Uptime Collapse GitHub Average Uptime Collapse Apple M1 Beats PC Chips Apple M1 Beats PC Chips Абсурдный мониторинг компьютера Абсурдный мониторинг компьютера GPT-4 Is Much Bigger Than GPT-3 GPT-4 Is Much Bigger Than GPT-3 Cat Beats Every AI Model Cat Beats Every AI Model 14900K Crashed Against 9950X 14900K Crashed Against 9950X