DROP

AI benchmark results and model performance

Type: percentageMath

Recorded Results

Average Score

64.54%

Score Range

28.60% - 91.60%

Leading Model (lowest score)

28.60% - Ernie 4.5 21B A3B

Scores Over Time

Individual benchmark scores plotted by date.

Models Using This Benchmark

Organisation	Model	Reported	Top Score	Info	Self Reported	Source
Baidu	Ernie 4.5 21B A3B	-	28.60%	inferred version-family alias from ernie-4.5	Yes	Source
Baidu	Ernie 4.5 21B A3B Thinking	-	28.60%	inferred version-family alias from ernie-4.5	Yes	Source
Baidu	Ernie 4.5 300B A47B	-	28.60%	inferred version-family alias from ernie-4.5	Yes	Source
Baidu	Ernie 4.5 Turbo	-	28.60%	inferred version-family alias from ernie-4.5	Yes	Source
Baidu	Ernie 4.5 VL 28B A3B	-	28.60%	inferred version-family alias from ernie-4.5	Yes	Source
Baidu	Ernie 4.5 VL 424B A47B	-	28.60%	inferred version-family alias from ernie-4.5	Yes	Source
IBM	Granite 4.0 Micro	02 Oct 2025	46.20%	inferred high-confidence family alias from granite-4.0-tiny-preview (score=0.4700; benches=12)	Yes	Source
IBM	Granite 4.0 Small	02 Oct 2025	46.20%	inferred high-confidence family alias from granite-4.0-tiny-preview (score=0.4700; benches=12)	Yes	Source
IBM	Granite 4.0 Tiny	02 Oct 2025	46.20%	inferred alias from granite-4.0-tiny-preview	Yes	Source
IBM	Granite 4.0 Tiny Preview	02 May 2025	46.20%	-	Yes	Source
Google	Gemma 3n E2B	25 Jun 2025	53.90%	-	Yes	Source
IBM	Granite 3.0 8B Instruct	-	59.36%	inferred high-confidence family alias from granite-3.3-8b-instruct (score=0.4911; benches=14)	Yes	Source
IBM	Granite 3.1 8B Instruct	-	59.36%	inferred high-confidence family alias from granite-3.3-8b-instruct (score=0.4911; benches=14)	Yes	Source
IBM	Granite 3.2 8B Instruct	-	59.36%	inferred high-confidence family alias from granite-3.3-8b-instruct (score=0.4911; benches=14)	Yes	Source
IBM	Granite 3.2 8B Instruct Preview	-	59.36%	inferred high-confidence family alias from granite-3.3-8b-instruct (score=0.4687; benches=14)	Yes	Source
IBM	Granite 3.3 2B Instruct	16 Apr 2025	59.36%	inferred family alias from granite-3.3-8b-instruct (score=0.3627; benches=14)	Yes	Source
IBM	Granite 3.3 8B Instruct	16 Apr 2025	59.36%	-	Yes	Source
IBM	Granite Guardian 3.0 8B	-	59.36%	inferred family alias from granite-3.3-8b-instruct (score=0.4062; benches=14)	Yes	Source
IBM	Granite Guardian 3.1 8B	-	59.36%	inferred family alias from granite-3.3-8b-instruct (score=0.4062; benches=14)	Yes	Source
IBM	Granite Guardian 3.3 8B	-	59.36%	inferred high-confidence family alias from granite-3.3-8b-instruct (score=0.5071; benches=14)	Yes	Source
IBM	Granite Speech 3.2 8B	-	59.36%	inferred family alias from granite-3.3-8b-instruct (score=0.4062; benches=14)	Yes	Source
IBM	Granite Speech 3.3 8B	-	59.36%	inferred high-confidence family alias from granite-3.3-8b-instruct (score=0.5071; benches=14)	Yes	Source
Google	Gemma 3n E4B	25 Jun 2025	60.80%	-	Yes	Source
OpenAI	GPT 3.5 Turbo 16K 0613	13 Jun 2023	70.20%	inferred family alias from gpt-3.5-turbo-0125 (score=0.4244; benches=8)	No	Source
OpenAI	GPT 3.5 Turbo (2023-11-06)	28 Sept 2023	70.20%	inferred modality/version alias from gpt-3.5-turbo-0125	No	Source
OpenAI	GPT 3.5 Turbo Instruct	06 Nov 2023	70.20%	inferred high-confidence family alias from gpt-3.5-turbo-0125 (score=0.5539; benches=8)	No	Source
Google	Gemini 1.5 Pro 001	23 May 2024	74.90%	inferred alias from gemini-1.5-pro	Yes	Source
Google	Gemini 1.5 Pro Exp (2024-08-01)	01 Aug 2024	74.90%	inferred alias from gemini-1.5-pro	Yes	Source
Google	Gemini 1.5 Pro Exp (2024-08-27)	27 Aug 2024	74.90%	inferred alias from gemini-1.5-pro	Yes	Source
Google	Gemini Robotics ER 1.5 Preview	25 Sept 2025	74.90%	inferred family alias from gemini-1.5-pro (score=0.3717; benches=23)	Yes	Source
Google	LearnLM 1.5 Pro Experimental	19 Nov 2024	74.90%	inferred family alias from gemini-1.5-pro (score=0.3700; benches=23)	Yes	Source
Microsoft	Phi 1	-	75.50%	inferred family alias from phi-4 (score=0.3100; benches=13)	Yes	Source
Microsoft	Phi 2	-	75.50%	inferred family alias from phi-4 (score=0.3100; benches=13)	Yes	Source
Microsoft	Phi 4	12 Dec 2024	75.50%	-	Yes	Source
Meituan	LongCat Flash Chat	-	79.06%	inferred high-confidence family alias from longcat-flash-chat (score=0.4667; benches=16)	Yes	Source
OpenAI	GPT 4o Mini Audio Preview	17 Dec 2024	79.70%	inferred modality/version alias from gpt-4o-mini-2024-07-18	Yes	Source
OpenAI	GPT 4o Mini Realtime Preview	17 Dec 2024	79.70%	inferred modality/version alias from gpt-4o-mini-2024-07-18	Yes	Source
OpenAI	GPT 4o Mini Search Preview	11 Mar 2025	79.70%	inferred modality/version alias from gpt-4o-mini-2024-07-18	Yes	Source
OpenAI	GPT 4o Mini Transcribe (2025-12-15)	15 Dec 2025	79.70%	inferred modality/version alias from gpt-4o-mini-2024-07-18	Yes	Source
OpenAI	GPT 4o Mini Transcribe (2025-03-20)	20 Mar 2025	79.70%	inferred modality/version alias from gpt-4o-mini-2024-07-18	Yes	Source
OpenAI	GPT 4o Mini TTS (2025-12-15)	15 Dec 2025	79.70%	inferred modality/version alias from gpt-4o-mini-2024-07-18	Yes	Source
OpenAI	GPT 4o Mini TTS (2025-03-20)	20 Mar 2025	79.70%	inferred modality/version alias from gpt-4o-mini-2024-07-18	Yes	Source
OpenAI	GPT 4 32K 0613	13 Jun 2023	80.90%	inferred high-confidence family alias from gpt-4-0613 (score=0.4899; benches=12)	Yes	Source
OpenAI	GPT 4o (2024-05-13)	13 May 2024	83.40%	-	Yes	Source
OpenAI	GPT 4 Turbo Preview (1106)	06 Nov 2023	86%	inferred version-family alias from gpt-4-turbo-2024-04-09	Yes	Source
DeepSeek	DeepSeek OCR	20 Oct 2025	91.60%	inferred family alias from deepseek-v3 (score=0.3000; benches=20)	Yes	Source
DeepSeek	DeepSeek V2 (2024-06-28)	28 Jun 2024	91.60%	inferred family alias from deepseek-v3 (score=0.4159; benches=20)	Yes	Source
DeepSeek	DeepSeek V4	-	91.60%	inferred high-confidence family alias from deepseek-v3 (score=0.5818; benches=20)	Yes	Source