GPQA Benchmark

GPQA Benchmark | Phaseo

Models Using This Benchmark

Organisation	Model	Reported	Top Score	Info	Self Reported	Source
Google	Gemini 3.1 Pro Preview	19 Feb 2026	94.30%	GPQA Diamond subset	Yes	Source
Google	Gemini 3.1 Pro Preview Customtools	19 Feb 2026	94.30%	inferred alias of gemini-3.1-pro-preview	Yes	Source
OpenAI	GPT 5 Pro	07 Aug 2025	92.80%	inferred family alias from gpt-5.4 (score=0.4083; benches=19)	Yes	Source
OpenAI	GPT 5 Search API	14 Oct 2025	92.80%	inferred family alias from gpt-5.4 (score=0.3050; benches=19)	Yes	Source
OpenAI	GPT 5.4	05 Mar 2026	92.80%	-	Yes	Source
OpenAI	GPT 5.2 Chat	11 Dec 2025	92.40%	inferred alias from gpt-5.2-2025-12-11	Yes	Source
Google	Gemini 3 Pro Image Preview (Nano Banana Pro)	20 Nov 2025	91.90%	inferred modality/version alias from gemini-3-pro-preview	Yes	Source
Google	Gemini 3 Pro Preview	18 Nov 2025	91.90%	GPQA Diamond subset	Yes	Source
Anthropic	Claude Opus 4.6	05 Feb 2026	91.30%	-	Yes	Source
Google	Gemini 3 Flash Preview	17 Dec 2025	90.40%	GPQA Diamond subset	Yes	Source
Qwen	Qwen 3.6 Plus	01 Apr 2026	90.40%	-	Yes	Source
Anthropic	Claude Sonnet 4.6	17 Feb 2026	89.90%	-	Yes	Source
ByteDance	Seed 2.0 Pro	14 Feb 2026	88.90%	-	Yes	Source
Qwen	Qwen 3.5 397B A17B	16 Feb 2026	88.40%	-	Yes	Source
SpaceXAI	Grok 4 Heavy	10 Jul 2025	88.40%	-	Yes	Source
OpenAI	GPT 5.1 Chat	13 Nov 2025	88.10%	inferred alias from gpt-5.1-2025-11-13	Yes	Source
OpenAI	GPT 5.4 Mini	17 Mar 2026	88%	-	Yes	Source
Moonshot	Kimi K2.5	27 Jan 2026	87.60%	-	Yes	Source
SpaceXAI	Grok 4	10 Jul 2025	87.50%	-	Yes	Source
Google	Gemini 3.1 Flash Image Preview (Nano Banana 2)	26 Feb 2026	86.90%	inferred modality/version alias from gemini-3.1-flash-lite-preview	Yes	Source
Google	Gemini 3.1 Flash-Lite	07 May 2026	86.90%	GPQA Diamond subset	Yes	Source
Google	Gemini 3.1 Flash Lite Preview	03 Mar 2026	86.90%	GPQA Diamond subset	Yes	Source
Google	Veo 3.1 Lite Preview	31 Mar 2026	86.90%	manual fallback alias from gemini-3.1-flash-lite-preview	Yes	Source
Qwen	Qwen 3.5 122B A10B	24 Feb 2026	86.60%	-	Yes	Source
Google	Gemini 2.5 Pro Preview TTS (2025-05-20)	20 May 2025	86.40%	inferred family alias from gemini-2.5-pro-preview-06-05 (score=0.4243; benches=13)	Yes	Source
z.AI	GLM 4.7	22 Dec 2025	85.70%	-	Yes	Source
Qwen	Qwen 3.5 27B	24 Feb 2026	85.50%	-	Yes	Source
Qwen	Qwen 3.5 Flash	23 Feb 2026	85.50%	inferred family alias from qwen3.5-27b (score=0.4147; benches=81)	Yes	Source
ByteDance	Seed 2.0 Lite	14 Feb 2026	85.10%	-	Yes	Source
Baidu	Ernie 5.0	22 Jan 2026	85%	GPQA Diamond	Yes	Source
Baidu	Ernie 5.0 0110	-	85%	inferred version-family alias from ernie-5.0	Yes	Source
Baidu	Ernie 5.0 Preview 1203	-	85%	inferred version-family alias from ernie-5.0	Yes	Source
Baidu	Ernie 5.0 Preview 1220	-	85%	inferred version-family alias from ernie-5.0	Yes	Source
SpaceXAI	Grok 3	18 Apr 2025	84.60%	-	Yes	Source
Moonshot	Kimi K2 Thinking	06 Nov 2025	84.50%	inferred alias from kimi-k2-thinking-0905	Yes	Source
Qwen	Qwen 3.5 35B A3B	24 Feb 2026	84.20%	-	Yes	Source
OpenAI	ChatGPT 4o	13 May 2024	84%	inferred alias from chatgpt-4o-latest	Yes	Source
SpaceXAI	Grok 3 Mini	18 Apr 2025	84%	-	Yes	Source
Xiaomi	MiMo V2 Flash	16 Dec 2025	83.70%	GPQA Diamond	Yes	Source
Google	Gemini 2.5 Computer Use Preview	07 Oct 2025	83%	inferred family alias from gemini-2.5-pro (score=0.3960; benches=16)	Yes	Source
Google	Gemini 2.5 Pro Experimental (2025-03-25)	25 Mar 2025	83%	inferred alias from gemini-2.5-pro	Yes	Source
Google	Gemini 2.5 Pro Preview TTS (2025-12-10)	10 Dec 2025	83%	inferred modality/version alias from gemini-2.5-pro	Yes	Source
Google	Gemini Embedding 2 Preview	10 Mar 2026	83%	manual fallback alias from gemini-2.5-pro	Yes	Source
Google	Gemini 2.5 Flash Exp Native Audio Thinking Dialog	-	82.80%	inferred modality/version alias from gemini-2.5-flash	Yes	Source
Google	Gemini 2.5 Flash Image (Nano Banana)	02 Oct 2025	82.80%	inferred modality/version alias from gemini-2.5-flash	Yes	Source
Google	Gemini 2.5 Flash Image Preview (Nano Banana)	25 Aug 2025	82.80%	inferred modality/version alias from gemini-2.5-flash	Yes	Source
Google	Gemini 2.5 Flash Native Audio Preview (2025-09-23)	-	82.80%	inferred modality/version alias from gemini-2.5-flash	Yes	Source
Google	Gemini 2.5 Flash Preview (2025-09-25)	25 Sept 2025	82.80%	inferred alias from gemini-2.5-flash	Yes	Source
Google	Gemini 2.5 Flash Preview Native Audio Dialog	-	82.80%	inferred modality/version alias from gemini-2.5-flash	Yes	Source
Google	Gemini 2.5 Flash Preview TTS (2025-05-20)	20 May 2025	82.80%	inferred modality/version alias from gemini-2.5-flash	Yes	Source
Google	Gemini 2.5 Flash Preview TTS (2025-12-10)	10 Dec 2025	82.80%	inferred modality/version alias from gemini-2.5-flash	Yes	Source
Google	Gemini Live 2.5 Flash Preview	09 Apr 2025	82.80%	inferred high-confidence family alias from gemini-2.5-flash (score=0.5083; benches=14)	Yes	Source
OpenAI	GPT 5.4 Nano	17 Mar 2026	82.80%	-	Yes	Source
Nvidia	Nemotron 3 Super	11 Mar 2026	82.70%	-	Yes	Source
Qwen	Qwen 3.5 9B	02 Mar 2026	81.70%	-	Yes	Source
OpenAI	o4 Mini	16 Apr 2025	81.40%	-	Yes	Source
OpenAI	o4 mini Deep Research	26 Jun 2025	81.40%	inferred modality/version alias from o4-mini	Yes	Source
Qwen	Qwen 3 235B A22B Thinking 2507	-	81.10%	-	Yes	Source
MiniMax	MiniMax M2.1	23 Dec 2025	81%	-	Yes	Source
z.AI	GLM 4.6	30 Sept 2025	81%	-	Yes	Source
OpenAI	GPT OSS 120b	05 Aug 2025	80.10%	-	Yes	Source
OpenAI	GPT OSS Safeguard 120b	29 Oct 2025	80.10%	inferred high-confidence family alias from gpt-oss-120b (score=0.5102; benches=7)	Yes	Source
DeepSeek	DeepSeek OCR 2	-	79.90%	inferred family alias from deepseek-v3.2-exp (score=0.3809; benches=14)	Yes	Source
DeepSeek	DeepSeek V3.2 Exp	29 Sept 2025	79.90%	-	Yes	Source
z.AI	GLM 4.5	28 Jul 2025	79.10%	-	Yes	Source
OpenAI	o1 pro	19 Mar 2025	79%	-	Yes	Source
MiniMax	MiniMax M2	27 Oct 2025	78%	-	Yes	-
MiniMax	MiniMax M2 Her	24 Jan 2026	78%	inferred modality/version alias from minimax-m2	Yes	-
OpenAI	o3 mini	30 Jan 2025	77.20%	-	Yes	Source
Qwen	Qwen 3 Next 80B A3B Thinking	-	77.20%	-	Yes	Source
Qwen	Qwen 3.5 4B	02 Mar 2026	76.20%	-	Yes	Source
Nvidia	Llama 3.1 Nemotron Ultra 253B v1	07 Apr 2025	76.01%	-	Yes	Source
z.AI	GLM 4.7 Flash	19 Jan 2026	75.20%	-	Yes	Source
Nvidia	Nemotron Nano 3 30B A3B	15 Dec 2025	75%	-	Yes	Source
z.AI	GLM 4.5 Air	28 Jul 2025	75%	-	Yes	Source
DeepSeek	DeepSeek V3.1	21 Aug 2025	74.90%	Non-thinking: 74.9%, Thinking: 80.1%	Yes	Source
DeepSeek	DeepSeek V3.1 Terminus	22 Sept 2025	74.90%	inferred alias from deepseek-v3.1	Yes	Source
Qwen	Qwen 3 VL 30B A3B Thinking	-	74.40%	-	Yes	-
Google	Gemini 2.0 Flash Thinking Exp (2024-12-19)	19 Dec 2024	74.20%	inferred alias from gemini-2.0-flash-thinking	Yes	Source
Google	Gemini 2.0 Flash Thinking Exp (2025-01-21)	21 Jan 2025	74.20%	inferred alias from gemini-2.0-flash-thinking	Yes	Source
Baidu	Ernie 4.5 21B A3B	-	74%	inferred version-family alias from ernie-4.5	Yes	Source
Baidu	Ernie 4.5 21B A3B Thinking	-	74%	inferred version-family alias from ernie-4.5	Yes	Source
Baidu	Ernie 4.5 300B A47B	-	74%	inferred version-family alias from ernie-4.5	Yes	Source
Baidu	Ernie 4.5 Turbo	-	74%	inferred version-family alias from ernie-4.5	Yes	Source
Baidu	Ernie 4.5 VL 28B A3B	-	74%	inferred version-family alias from ernie-4.5	Yes	Source
Baidu	Ernie 4.5 VL 424B A47B	-	74%	inferred version-family alias from ernie-4.5	Yes	Source
Inception	Mercury 2	24 Feb 2026	74%	-	Yes	Source
OpenAI	o1 preview	12 Sept 2024	73.30%	-	Yes	Source
Meituan	LongCat Flash Chat	-	73.23%	inferred high-confidence family alias from longcat-flash-chat (score=0.4667; benches=16)	Yes	Source
Qwen	Qwen 3 VL 32B Thinking	-	73.10%	-	Yes	-
Anthropic	Claude Haiku 4.5	15 Oct 2025	73%	inferred alias from claude-haiku-4-5-20251001	Yes	Source
Qwen	Qwen 3 Next 80B A3B Instruct	-	72.90%	-	Yes	Source
OpenAI	GPT OSS 20b	05 Aug 2025	71.50%	-	Yes	Source
OpenAI	GPT OSS Safeguard 20b	29 Oct 2025	71.50%	inferred high-confidence family alias from gpt-oss-20b (score=0.5137; benches=7)	Yes	Source
Mistral	Mistral Small 1.0	26 Feb 2024	71.20%	inferred family alias from mistral-small-latest (score=0.3650; benches=9)	Yes	Source
Mistral	Mistral Small 2.0	17 Sept 2024	71.20%	inferred family alias from mistral-small-latest (score=0.3650; benches=9)	Yes	Source
Mistral	Mistral Small 4	16 Mar 2026	71.20%	-	Yes	Source
Mistral	Mistral Small Creative	16 Dec 2025	71.20%	inferred family alias from mistral-small-latest (score=0.4273; benches=9)	Yes	Source
Mistral	Magistral Medium 1.1	24 Jul 2025	70.80%	inferred version-family alias from magistral-medium	Yes	Source
Mistral	Magistral Medium 1.2	17 Sept 2025	70.80%	inferred version-family alias from magistral-medium	Yes	Source
Qwen	Qwen 3 VL 30B A3B Instruct	-	70.40%	-	Yes	-
OpenAI	GPT 4o (2024-08-06)	06 Aug 2024	70.10%	-	Yes	Source
OpenAI	GPT 4o Audio (2025-06-03)	03 Jun 2025	70.10%	inferred modality/version alias from gpt-4o-2024-08-06	Yes	Source
OpenAI	GPT 4o Audio (2024-10-01)	01 Oct 2024	70.10%	inferred modality/version alias from gpt-4o-2024-08-06	Yes	Source
OpenAI	GPT 4o Audio (2024-12-17)	17 Dec 2024	70.10%	inferred modality/version alias from gpt-4o-2024-08-06	Yes	Source
OpenAI	GPT 4o Realtime Preview (2025-06-03)	03 Jun 2025	70.10%	inferred modality/version alias from gpt-4o-2024-08-06	Yes	Source
OpenAI	GPT 4o Realtime Preview (2024-10-01)	01 Oct 2024	70.10%	inferred modality/version alias from gpt-4o-2024-08-06	Yes	Source
OpenAI	GPT 4o Search Preview	11 Mar 2025	70.10%	inferred modality/version alias from gpt-4o-2024-08-06	Yes	Source
OpenAI	GPT 4o Transcribe	20 Mar 2025	70.10%	inferred modality/version alias from gpt-4o-2024-08-06	Yes	Source
OpenAI	GPT 4o Transcribe Diarize	15 Oct 2025	70.10%	inferred modality/version alias from gpt-4o-2024-08-06	Yes	Source
MiniMax	MiniMax M1 80K	16 Jun 2025	70%	-	Yes	-
Qwen	Qwen 3 8B	-	69.90%	inferred high-confidence family alias from qwen3-vl-8b-thinking (score=0.4600; benches=50)	Yes	-
Qwen	Qwen 3 Embedding 8B	-	69.90%	inferred family alias from qwen3-vl-8b-thinking (score=0.3850; benches=50)	Yes	-
Qwen	Qwen 3 Guard Gen 8B	-	69.90%	inferred family alias from qwen3-vl-8b-thinking (score=0.3400; benches=50)	Yes	-
Qwen	Qwen 3 Guard Stream 8B	-	69.90%	inferred family alias from qwen3-vl-8b-thinking (score=0.3371; benches=50)	Yes	-
Qwen	Qwen 3 Reranker 8B	-	69.90%	inferred family alias from qwen3-vl-8b-thinking (score=0.3850; benches=50)	Yes	-
Qwen	Qwen 3 VL 8B Thinking	-	69.90%	-	Yes	-
Qwen	Qwen 3 VL Embedding 8B	-	69.90%	inferred high-confidence family alias from qwen3-vl-8b-thinking (score=0.5232; benches=50)	Yes	-
Qwen	Qwen 3 VL Reranker 8B	-	69.90%	inferred high-confidence family alias from qwen3-vl-8b-thinking (score=0.5275; benches=50)	Yes	-
Meta	Llama 4 Maverick	05 Apr 2025	69.80%	-	Yes	Source
OpenAI	GPT 4.5	27 Feb 2025	69.50%	-	Yes	Source
MiniMax	Minimax M1 40K	16 Jun 2025	69.20%	-	Yes	-
Microsoft	Phi 4 Reasoning Plus	30 Apr 2025	68.90%	-	Yes	Source
Qwen	Qwen 3 VL 32B Instruct	-	68.90%	-	Yes	-
Mistral	Magistral Small 1.1	24 Jul 2025	68.18%	inferred version-family alias from magistral-small-2506	Yes	Source
Mistral	Magistral Small 1.2	17 Sept 2025	68.18%	inferred version-family alias from magistral-small-2506	Yes	Source
Mistral	Ministral 8B	09 Oct 2024	66.80%	inferred alias from ministral-8b-latest	Yes	Source
Nvidia	Llama 3.3 Nemotron Super 49B v1	18 Mar 2025	66.67%	-	Yes	Source
Nvidia	Llama 3.3 Nemotron Super 49B V1.5	-	66.67%	inferred version-family alias from llama-3.3-nemotron-super-49b-v1	Yes	Source
Microsoft	Phi 4 Reasoning	30 Apr 2025	65.80%	-	Yes	Source
Qwen	Qwen 3 30B A3B	-	65.80%	-	Yes	Source
Qwen	Qwen 3 30B A3B Instruct 2507	-	65.80%	inferred version-family alias from qwen3-30b-a3b	Yes	Source
Qwen	Qwen 3 30B A3B Thinking 2507	-	65.80%	inferred version-family alias from qwen3-30b-a3b	Yes	Source
Qwen	Qwen 3 Coder 30B A3B Instruct	-	65.80%	inferred high-confidence family alias from qwen3-30b-a3b (score=0.5007; benches=8)	Yes	Source
Qwen	Qwen 3 Omni 30B A3B Captioner	-	65.80%	inferred family alias from qwen3-30b-a3b (score=0.4129; benches=8)	Yes	Source
Qwen	Qwen 3 Omni 30B A3B Instruct	-	65.80%	inferred high-confidence family alias from qwen3-30b-a3b (score=0.4819; benches=8)	Yes	Source
Qwen	Qwen 3 Omni 30B A3B Thinking	-	65.80%	inferred high-confidence family alias from qwen3-30b-a3b (score=0.4819; benches=8)	Yes	Source
Qwen	QwQ 32B	-	65.20%	-	Yes	Source
Qwen	QwQ 32B Preview	-	65.20%	-	Yes	Source
Google	Gemini 2.5 Flash Lite Preview (2025-09-25)	25 Sept 2025	64.60%	inferred alias from gemini-2.5-flash-lite	Yes	Source
Qwen	Qwen 3 4B Instruct 2507	-	64.10%	inferred family alias from qwen3-vl-4b-thinking (score=0.3462; benches=48)	Yes	-
Qwen	Qwen 3 4B SafeRL	-	64.10%	inferred family alias from qwen3-vl-4b-thinking (score=0.3850; benches=48)	Yes	-
Qwen	Qwen 3 4B Thinking 2507	-	64.10%	inferred family alias from qwen3-vl-4b-thinking (score=0.3462; benches=48)	Yes	-
Qwen	Qwen 3 Embedding 4B	-	64.10%	inferred family alias from qwen3-vl-4b-thinking (score=0.3850; benches=48)	Yes	-
Qwen	Qwen 3 Guard Gen 4B	-	64.10%	inferred family alias from qwen3-vl-4b-thinking (score=0.3400; benches=48)	Yes	-
Qwen	Qwen 3 Guard Stream 4B	-	64.10%	inferred family alias from qwen3-vl-4b-thinking (score=0.3371; benches=48)	Yes	-
Qwen	Qwen 3 Reranker 4B	-	64.10%	inferred family alias from qwen3-vl-4b-thinking (score=0.3850; benches=48)	Yes	-
Qwen	Qwen 3 VL 4B Thinking	-	64.10%	-	Yes	-
Nvidia	Nvidia Nemotron Nano 12B V2	-	64%	inferred high-confidence family alias from nvidia-nemotron-nano-9b-v2 (score=0.4889; benches=6)	Yes	Source
Nvidia	Nvidia Nemotron Nano 9B V2	-	64%	-	Yes	Source
Google	Gemini 2.0 Flash	05 Feb 2025	62.10%	-	Yes	Source
Google	Gemini 2.0 Flash Exp	-	62.10%	inferred alias from gemini-2.0-flash	Yes	Source
Google	Gemini 2.0 Pro Exp (2025-02-05)	05 Feb 2025	62.10%	inferred modality/version alias from gemini-2.0-flash	Yes	Source
Google	LearnLM 2.0 Flash Experimental	17 Apr 2025	62.10%	inferred family alias from gemini-2.0-flash (score=0.3700; benches=13)	Yes	Source
Qwen	Qwen 3 14B	-	62%	inferred family alias from qwen3-max (score=0.3333; benches=6)	Yes	Source
Qwen	Qwen 3 Max Thinking	26 Jan 2026	62%	inferred alias from qwen3-max	Yes	Source
Qwen	Qwen 3 TTS (2025-11-27)	-	62%	inferred family alias from qwen3-max (score=0.3833; benches=6)	Yes	Source
OpenAI	o1 mini	12 Sept 2024	60%	-	Yes	Source
DeepSeek	DeepSeek OCR	20 Oct 2025	59.10%	inferred family alias from deepseek-v3 (score=0.3000; benches=20)	Yes	Source
DeepSeek	DeepSeek V2 (2024-06-28)	28 Jun 2024	59.10%	inferred family alias from deepseek-v3 (score=0.4159; benches=20)	Yes	Source
DeepSeek	DeepSeek V4	-	59.10%	inferred high-confidence family alias from deepseek-v3 (score=0.5818; benches=20)	Yes	Source
Google	Gemini 1.5 Pro 001	23 May 2024	59.10%	inferred alias from gemini-1.5-pro	Yes	Source
Google	Gemini 1.5 Pro Exp (2024-08-01)	01 Aug 2024	59.10%	inferred alias from gemini-1.5-pro	Yes	Source
Google	Gemini 1.5 Pro Exp (2024-08-27)	27 Aug 2024	59.10%	inferred alias from gemini-1.5-pro	Yes	Source
Google	Gemini Robotics ER 1.5 Preview	25 Sept 2025	59.10%	inferred family alias from gemini-1.5-pro (score=0.3717; benches=23)	Yes	Source
Google	LearnLM 1.5 Pro Experimental	19 Nov 2024	59.10%	inferred family alias from gemini-1.5-pro (score=0.3700; benches=23)	Yes	Source
Meta	Llama 4 Scout	05 Apr 2025	57.20%	-	Yes	Source
Microsoft	Phi 1	-	56.10%	inferred family alias from phi-4 (score=0.3100; benches=13)	Yes	Source
Microsoft	Phi 2	-	56.10%	inferred family alias from phi-4 (score=0.3100; benches=13)	Yes	Source
Microsoft	Phi 4	12 Dec 2024	56.10%	-	Yes	Source
SpaceXAI	Grok 2	13 Aug 2024	56%	-	Yes	Source
Nvidia	Llama 3.1 Nemotron Nano 4B V1.1	-	54.10%	inferred high-confidence family alias from llama-3.1-nemotron-nano-8b-v1 (score=0.5523; benches=7)	Yes	Source
Nvidia	Llama 3.1 Nemotron Nano 8B V1	18 Mar 2025	54.10%	-	Yes	Source
OpenAI	GPT 4o (2024-05-13)	13 May 2024	53.60%	-	Yes	Source
Mistral	Ministral 3B	09 Oct 2024	53.40%	inferred alias from ministral-3b-latest	Yes	Source
Microsoft	Phi 4 Mini Flash Reasoning	-	52%	inferred modality/version alias from phi-4-mini-reasoning	Yes	Source
Microsoft	Phi 4 Mini Reasoning	30 Apr 2025	52%	-	Yes	Source
Qwen	Qwen 3.5 2B	02 Mar 2026	51.60%	-	Yes	Source
Google	Gemini 2.0 Flash Lite	05 Feb 2025	51.50%	-	Yes	Source
Google	Gemini 1.5 Flash 001	23 May 2024	51%	inferred alias from gemini-1.5-flash	Yes	Source
Google	Gemini 1.5 Flash Preview	14 May 2024	51%	inferred alias from gemini-1.5-flash	Yes	Source
SpaceXAI	Grok 2 Mini	13 Aug 2024	51%	-	Yes	Source
Meta	Llama 2 70B Chat	20 Jun 2023	50.50%	inferred family alias from llama-3.3-70b-instruct (score=0.3129; benches=9)	Yes	Source
Qwen	Qwen 72B	-	49%	inferred family alias from qwen-2.5-72b-instruct (score=0.3060; benches=14)	Yes	Source
OpenAI	GPT 4 Turbo (2023-11-06)	06 Nov 2023	48%	inferred version-family alias from gpt-4-turbo-2024-04-09	Yes	Source
Qwen	Qwen 3 235B A22B	-	47.47%	-	Yes	Source
Meta	Llama 3.2 90B Vision Instruct	-	46.70%	inferred high-confidence family alias from llama-3.2-90b-instruct (score=0.5873; benches=13)	Yes	Source
Qwen	Qwen 2.5 Coder 32B Instruct	-	46%	inferred high-confidence family alias from qwen2.5-vl-32b (score=0.4641; benches=28)	Yes	Source
Qwen	Qwen 2.5 VL 32B Instruct	24 Mar 2025	46%	-	Yes	Source
Qwen	Qwen 2.5 VL 3B Instruct	-	46%	inferred high-confidence family alias from qwen2.5-vl-32b (score=0.4914; benches=28)	Yes	Source
IBM	Granite 4.1 30B	29 Apr 2026	45.76%	0-shot, CoT	Yes	Source
Qwen	Qwen 14B	-	45.50%	inferred family alias from qwen-2.5-14b-instruct (score=0.3060; benches=16)	Yes	Source
Mistral	Mistral Small 3.2	20 Jun 2025	44.22%	5 Shot CoT	Yes	Source
Mistral	Mistral Large 1.0	26 Feb 2024	43.90%	inferred family alias from mistral-large-latest (score=0.3650; benches=5)	Yes	Source
Qwen	Qwen 2 Math 72B	-	42.40%	inferred high-confidence family alias from qwen2-72b-instruct (score=0.4667; benches=17)	Yes	Source
Qwen	Qwen 2 Math RM 72B	-	42.40%	inferred family alias from qwen2-72b-instruct (score=0.3917; benches=17)	Yes	Source
IBM	Granite 4.1 8B	29 Apr 2026	41.96%	0-shot, CoT	Yes	Source
Google	Gemini Diffusion	20 May 2025	40.40%	-	Yes	Source
OpenAI	GPT 4o Mini Audio Preview	17 Dec 2024	40.20%	inferred modality/version alias from gpt-4o-mini-2024-07-18	Yes	Source
OpenAI	GPT 4o Mini Realtime Preview	17 Dec 2024	40.20%	inferred modality/version alias from gpt-4o-mini-2024-07-18	Yes	Source
OpenAI	GPT 4o Mini Search Preview	11 Mar 2025	40.20%	inferred modality/version alias from gpt-4o-mini-2024-07-18	Yes	Source
OpenAI	GPT 4o Mini Transcribe (2025-12-15)	15 Dec 2025	40.20%	inferred modality/version alias from gpt-4o-mini-2024-07-18	Yes	Source
OpenAI	GPT 4o Mini Transcribe (2025-03-20)	20 Mar 2025	40.20%	inferred modality/version alias from gpt-4o-mini-2024-07-18	Yes	Source
OpenAI	GPT 4o Mini TTS (2025-12-15)	15 Dec 2025	40.20%	inferred modality/version alias from gpt-4o-mini-2024-07-18	Yes	Source
OpenAI	GPT 4o Mini TTS (2025-03-20)	20 Mar 2025	40.20%	inferred modality/version alias from gpt-4o-mini-2024-07-18	Yes	Source
Google	Gemini 1.5 Flash 8B	15 Mar 2024	38.40%	-	Yes	Source
Google	Gemini 1.5 Flash 8B Exp (2024-08-27)	27 Aug 2024	38.40%	inferred alias from gemini-1.5-flash-8b	Yes	Source
Google	Gemini 1.5 Flash 8B Exp (2024-09-24)	24 Sept 2024	38.40%	inferred alias from gemini-1.5-flash-8b	Yes	Source
Microsoft	Phi 3.5 MoE instruct	23 Aug 2024	36.80%	-	Yes	Source
Qwen	Qwen 7B	-	36.40%	inferred family alias from qwen-2.5-7b-instruct (score=0.3083; benches=14)	Yes	Source
SpaceXAI	Grok 1.5	28 Mar 2024	35.90%	-	Yes	Source
OpenAI	GPT 4 32K 0613	13 Jun 2023	35.70%	inferred high-confidence family alias from gpt-4-0613 (score=0.4899; benches=12)	Yes	Source
Meta	Llama 3.2 11B Vision Instruct	-	32.80%	inferred high-confidence family alias from llama-3.2-11b-instruct (score=0.5873; benches=11)	Yes	Source
Meta	Llama 3.2 1B Instruct	25 Sept 2024	32.80%	inferred family alias from llama-3.2-11b-instruct (score=0.4200; benches=11)	Yes	Source
IBM	Granite 4.1 3B	29 Apr 2026	31.70%	0-shot, CoT	Yes	Source
OpenAI	GPT 3.5 Turbo 16K 0613	13 Jun 2023	30.80%	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	30.80%	inferred modality/version alias from gpt-3.5-turbo-0125	No	Source
OpenAI	GPT 3.5 Turbo Instruct	06 Nov 2023	30.80%	inferred high-confidence family alias from gpt-3.5-turbo-0125 (score=0.5539; benches=8)	No	Source
Qwen	Qwen 2.5 Coder 3B	-	30.80%	inferred family alias from qwen2.5-omni-7b (score=0.3000; benches=45)	Yes	Source
Qwen	Qwen 2.5 Coder 7B	-	30.80%	inferred high-confidence family alias from qwen2.5-omni-7b (score=0.4700; benches=45)	Yes	Source
Qwen	Qwen 2.5 Math 7B	-	30.80%	inferred high-confidence family alias from qwen2.5-omni-7b (score=0.4767; benches=45)	Yes	Source
Qwen	Qwen 2.5 Math 7B PRM800K	-	30.80%	inferred family alias from qwen2.5-omni-7b (score=0.3696; benches=45)	Yes	Source
Qwen	Qwen 2.5 Math PRM 7B	-	30.80%	inferred family alias from qwen2.5-omni-7b (score=0.4092; benches=45)	Yes	Source
Qwen	Qwen 2.5 Omni 3B	-	30.80%	inferred high-confidence family alias from qwen2.5-omni-7b (score=0.4933; benches=45)	Yes	Source
Qwen	Qwen 2.5 Omni 7B	-	30.80%	-	Yes	Source
Microsoft	Phi 3 Mini 128K Instruct	-	30.40%	inferred family alias from phi-3.5-mini-instruct (score=0.3533; benches=31)	Yes	Source
Microsoft	Phi 3.5 mini instruct	23 Aug 2024	30.40%	-	Yes	Source
Google	Gemini 1.0 Nano	06 Dec 2023	27.90%	inferred modality/version alias from gemini-1.0-pro	No	Source
Google	Gemini 1.0 Pro	06 Dec 2023	27.90%	-	No	Source
Google	Gemini 1.0 Pro Vision 001	15 Feb 2024	27.90%	inferred modality/version alias from gemini-1.0-pro	No	Source
Qwen	Qwen 2 Audio 7B	-	25.30%	inferred modality/version alias from qwen2-7b-instruct	Yes	Source
Qwen	Qwen 2 Math 7B	-	25.30%	inferred high-confidence family alias from qwen2-7b-instruct (score=0.4706; benches=14)	Yes	Source
Microsoft	Phi 4 Mini	01 Feb 2025	25.20%	-	Yes	Source
Qwen	Qwen 3.5 0.8B	02 Mar 2026	11.90%	-	Yes	Source

Recorded Results

Average Score

Score Range

GPQA

Recorded Results

Average Score

Score Range

Leading Model

Models Using This Benchmark