Benchmarks

AI model performance rankings

Click any two models to compare them side by side

		Index Scores			Benchmarks							Speed
#
1	Claude Opus 4.7 (Adaptive Reasoning, Max Effort)Anthropic	57.3	52.5	N/A	N/A	91.4	39.6	N/A	54.5	N/A	N/A	52
2	GPT-5.4 (xhigh)OpenAI	57.2	57.3	N/A	N/A	92.0	41.6	N/A	56.6	N/A	N/A	32
3	Gemini 3.1 Pro PreviewGooglevia Google AI Studio	57.2	55.5	N/A	N/A	94.1	44.7	N/A	58.9	N/A	N/A	71
4	Claude Opus 4.6 (Adaptive Reasoning, Max Effort)Anthropic	53.0	48.1	N/A	N/A	89.6	36.7	N/A	51.9	N/A	N/A	36
5	Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort)Anthropic	51.7	50.9	N/A	N/A	87.5	30.0	N/A	46.8	N/A	N/A	42
6	GLM-5.1 (Reasoning)Z AIvia Z.AI	51.4	43.4	N/A	N/A	86.8	28.0	N/A	43.8	N/A	N/A	21
7	Qwen3.6 PlusAlibaba	50.0	42.9	N/A	N/A	88.2	25.7	N/A	40.7	N/A	N/A	44
8	GLM-5 (Reasoning)Z AIvia Z.AI	49.8	44.2	N/A	N/A	82.0	27.2	N/A	46.2	N/A	N/A	32
9	GPT-5.4 mini (xhigh)OpenAI	48.1	51.5	N/A	N/A	87.5	26.6	N/A	49.9	N/A	N/A	83
10	Gemini 3 Flash Preview (Reasoning)Google	46.4	42.6	97.0	89.0	89.8	34.7	90.8	50.6	N/A	N/A	68
11	Qwen3.5 397B A17B (Reasoning)Alibaba	45.0	41.3	N/A	N/A	89.3	27.3	N/A	42.0	N/A	N/A	53
12	GPT-5.4 nano (xhigh)OpenAI	44.4	43.9	N/A	N/A	81.7	26.5	N/A	46.9	N/A	N/A	65
13	MiMo-V2-Flash (Feb 2026)Xiaomi	41.5	33.5	N/A	N/A	83.5	20.0	N/A	38.3	N/A	N/A	35
14	Grok 4xAI	41.5	40.5	92.7	86.6	87.7	23.9	81.9	45.7	99.0	94.3	45
15	Gemma 4 31B (Reasoning)Googlevia DeepInfra	39.2	38.7	N/A	N/A	85.7	22.7	N/A	43.4	N/A	N/A	32
16	Grok 4.1 Fast (Reasoning)xAI	38.6	30.9	89.3	85.4	85.3	17.6	82.2	44.2	N/A	N/A	116
17	Claude 4.5 Haiku (Reasoning)Anthropic	37.1	32.6	83.7	76.0	67.2	9.7	61.5	43.3	N/A	N/A	66
18	NVIDIA Nemotron 3 Super 120B A12B (Reasoning)NVIDIAvia Nebius	36.0	31.2	N/A	N/A	80.0	19.2	N/A	36.0	N/A	N/A	89
19	Grok 4 Fast (Reasoning)xAI	35.1	27.4	89.7	85.0	84.7	17.0	83.2	44.2	N/A	N/A	123
20	Gemini 3.1 Flash-Lite PreviewGoogle	33.5	30.1	N/A	N/A	82.2	16.2	N/A	41.9	N/A	N/A	44
21	gpt-oss-120B (high)OpenAIvia Groq	33.3	28.6	93.4	80.8	78.2	18.5	87.8	38.9	N/A	N/A	349
22	gpt-oss-120B (high)OpenAIvia Google	33.3	28.6	93.4	80.8	78.2	18.5	87.8	38.9	N/A	N/A	218
23	gpt-oss-120B (high)OpenAIvia Cerebras	33.3	28.6	93.4	80.8	78.2	18.5	87.8	38.9	N/A	N/A	768
24	GPT-4.1OpenAI	26.3	21.8	34.7	80.6	66.6	4.6	45.7	38.1	91.3	43.7	44
25	GPT-4.1 miniOpenAI	22.9	18.5	46.3	78.1	66.4	4.6	48.3	40.4	92.5	43.0	51
26	GPT-4o (Aug '24)OpenAI	18.6	16.6	N/A	N/A	52.1	2.9	31.7	33.1	79.5	11.7	14
27	DeepSeek R1 Distill Qwen 32BDeepSeekvia NextBit	17.2	N/A	63.0	73.9	61.5	5.5	27.0	37.6	94.1	68.7	24
28	DeepSeek R1 Distill Llama 70BDeepSeekvia DeepInfra	16.0	11.4	53.7	79.5	40.2	6.1	26.6	31.2	93.5	67.0	42
29	Gemini 2.0 Flash-Lite (Feb '25)Google	14.7	N/A	N/A	72.4	53.5	3.6	18.5	25.0	87.3	27.7	22
30	Llama 3.3 Instruct 70BMetavia Groq	14.5	10.7	7.7	71.3	49.8	4.0	28.8	26.0	77.3	30.0	144
31	Llama 4 ScoutMetavia Groq	13.5	6.7	14.0	75.2	58.7	4.3	29.9	17.0	84.4	28.3	100
32	GPT-4.1 nanoOpenAI	13.0	11.2	24.0	65.7	51.2	3.9	32.6	25.9	84.8	23.7	35
33	GPT-4o miniOpenAI	12.6	N/A	14.7	64.8	42.6	4.0	23.4	22.9	78.9	11.7	30

crafted by bart stefanski

vs...