Artificial Analysis LLM Performance Leaderboard

Independent performance benchmarks & pricing across API providers of LLMs. Definitions are below the table.

For further analysis and methodology, see artificialanalysis.ai.

Filters

Frontier Models:

Open Weights:

Size Class:

Reasoning:

Model Status:

	Features	Model Intelligence	Price	Output tokens/s	Latency	End-to-End Response Time
								Further Analysis
GPT-5 (high)	400k	69	$3.44	143.2	63.76	67.25	N/A
GPT-5 (high)	272k	69	$3.44	218.3	42.87	45.16	N/A
GPT-5 (medium)	400k	68	$3.44	179.7	35.26	38.04	N/A
GPT-5 (medium)	272k	68	$3.44	216.9	23.40	25.71	N/A
Grok 4	256k	68	$6.00	48.6	6.53	16.81	N/A
o3	200k	67	$3.50	218.8	11.05	13.34	N/A
o3	200k	67	$3.50	85.6	28.26	34.10	N/A
o4-mini (high)	200k	65	$1.93	107.3	48.44	53.10	N/A
o4-mini (high)	200k	65	$1.93	160.4	23.75	26.86	N/A
Gemini 2.5 Pro (AI_Studio)	1m	65	$3.44	140.2	30.36	33.93	N/A
Gemini 2.5 Pro Vertex	1m	65	$3.44	140.3	34.26	37.82	N/A
GPT-5 mini (medium)	400k	64	$0.69	73.6	29.17	35.96	N/A
GPT-5 mini (medium)	400k	64	$0.69	146.3	14.50	17.92	N/A
Qwen3 235B 2507 (Reasoning)	256k	64	$1.24	66.7	0.43	37.91	29.99
Qwen3 235B 2507 (Reasoning)	131k	64	$0.75	1,490.2	0.26	1.94	1.34
Qwen3 235B 2507 (Reasoning)	262k	64	$0.39	114.1	0.65	22.55	17.52
Qwen3 235B 2507 (Reasoning) (FP8)	262k	64	$0.25	28.5	0.59	88.28	70.15
Qwen3 235B 2507 (Reasoning)	131k	64	$0.97	49.7	0.80	51.09	40.23
Qwen3 235B 2507 (Reasoning) (FP8)	131k	64	$1.20	66.3	0.53	38.24	30.17
Qwen3 235B 2507 (Reasoning)	262k	64	$1.24	60.0	0.31	41.98	33.34
GPT-5 (low)	400k	63	$3.44	184.8	14.17	16.88	N/A
gpt-oss-120B (high)	131k	61	$0.26	169.9	0.46	15.18	11.77
gpt-oss-120B (high)	131k	61	$0.36	3,188.1	0.27	1.05	0.63
gpt-oss-120B (high)	131k	61	$0.26	223.4	0.38	11.57	8.95
gpt-oss-120B (high) Base	128k	61	$0.26	193.2	0.54	13.48	10.35
gpt-oss-120B (high) Vertex	131k	61	$0.26	230.0	0.18	11.05	8.70
gpt-oss-120B (high)	131k	61	$0.10	193.0	0.21	13.16	10.36
gpt-oss-120B (high)	131k	61	$0.26	270.3	36.92	46.17	7.40
gpt-oss-120B (high)	131k	61	$0.26	307.2	0.44	8.58	6.51
gpt-oss-120B (high)	131k	61	$0.18	173.0	0.23	14.68	11.56
gpt-oss-120B (high)	131k	61	$0.20	194.1	0.68	13.56	10.30
gpt-oss-120B (high)	131k	61	$0.26	228.9	33.58	44.51	8.74
gpt-oss-120B (high)	131k	61	$0.30	498.8	0.19	5.21	4.01
gpt-oss-120B (high)	131k	61	$0.26	333.7	43.10	50.59	5.99
gpt-oss-120B (high)	128k	61	$0.45	125.3	105.76	125.72	15.97
DeepSeek V3.1 (Reasoning)	128k	60	$0.96	19.4	2.95	131.61	102.92
DeepSeek V3.1 (Reasoning) (FP8)	164k	60	$0.90	22.4	1.09	112.71	89.30
DeepSeek V3.1 (Reasoning)	33k	60	$3.38	169.7	1.63	16.37	11.79
Claude 4 Sonnet Thinking	1m	59	$6.00	48.6	1.20	52.65	41.16
Claude 4 Sonnet Thinking Vertex	1m	59	$6.00	50.1	1.44	51.38	39.95
Claude 4 Sonnet Thinking	1m	59	$6.00	44.9	1.78	57.43	44.52
DeepSeek R1 0528	164k	59	$0.92	39.8	0.40	63.14	50.19
DeepSeek R1 0528	64k	59	$0.96	19.1	2.82	133.76	104.75
DeepSeek R1 0528	164k	59	$1.59	60.9	0.48	41.52	32.84
DeepSeek R1 0528	164k	59	$3.00	90.4	2.46	30.12	22.13
DeepSeek R1 0528	164k	59	$1.20	23.6	0.64	106.47	84.66
DeepSeek R1 0528 Fast	164k	59	$3.00	281.5	1.07	9.95	7.10
DeepSeek R1 0528 (Vertex)	164k	59	$2.36	213.9	0.50	12.19	9.35
DeepSeek R1 0528	128k	59	$0.53	48.8	0.26	51.47	40.97
DeepSeek R1 0528	128k	59	$2.36	76.9	0.54	33.06	26.02
DeepSeek R1 0528 Fast	164k	59	$4.25	231.6	0.53	11.32	8.64
DeepSeek R1 0528	164k	59	$0.91	80.4	0.37	31.45	24.87
DeepSeek R1 0528	164k	59	$1.15	34.1	0.60	73.95	58.68
DeepSeek R1 0528	131k	59	$1.18	112.0	0.51	22.83	17.86
DeepSeek R1 0528	33k	59	$5.50	175.6	1.92	16.15	11.39
DeepSeek R1 0528	164k	59	$4.00	273.9	0.82	9.95	7.30
DeepSeek R1 0528 (Throughput)	164k	59	$0.96	47.6	0.73	53.30	42.06
Gemini 2.5 Flash (Reasoning) (AI_Studio)	1m	58	$0.85	253.0	15.70	17.68	N/A
Gemini 2.5 Flash (Reasoning) (Vertex)	1m	58	$0.85	304.8	13.30	14.94	N/A
Grok 3 mini Reasoning (high)	131k	58	$0.35	186.8	0.58	13.96	10.71
Grok 3 mini Reasoning (high) Fast	131k	58	$1.45	187.9	0.69	14.00	10.64
Grok 3 mini Reasoning (high)	32k	58	$0.00	168.1	0.41	15.28	11.90
GLM-4.5 (FP8)	131k	56	$0.97	38.6	0.56	65.24	51.75
GLM-4.5	131k	56	$0.91	58.9	0.38	42.85	33.97
Claude 4 Opus Thinking	200k	55	$30.00	18.0	2.84	141.90	111.25
Claude 4 Opus Thinking Vertex	200k	55	$30.00	42.4	1.73	60.67	47.16
Claude 4 Opus Thinking	200k	55	$30.00	43.4	1.74	59.30	46.05
Qwen3 30B 2507 (Reasoning)	262k	54	$0.15	135.4	0.62	19.07	14.77
Qwen3 30B 2507 (Reasoning) (FP8)	262k	54	$0.90	184.3	0.46	14.02	10.85
GPT-5 nano (medium)	400k	54	$0.14	186.3	32.47	35.15	N/A
GPT-5 nano (medium)	400k	54	$0.14	253.3	25.85	27.82	N/A
GLM-4.5-Air	128k	53	$0.32	93.3	1.31	28.11	21.44
GLM-4.5-Air	131k	53	$0.42	157.4	0.24	16.13	12.71
GLM-4.5-Air (FP8)	131k	53	$0.42	201.4	0.37	12.78	9.93
Qwen3 235B 2507 (Non-reasoning)	262k	51	$0.33	57.2	0.42	9.17	N/A
Qwen3 235B 2507 (Non-reasoning)	131k	51	$0.75	1,288.0	0.32	0.71	N/A
Qwen3 235B 2507 (Non-reasoning)	262k	51	$2.00	59.7	2.09	10.46	N/A
Qwen3 235B 2507 (Non-reasoning)	262k	51	$0.30	59.4	0.65	9.06	N/A
Qwen3 235B 2507 (Non-reasoning) Vertex	256k	51	$0.44	84.4	0.43	6.36	N/A
Qwen3 235B 2507 (Non-reasoning) (FP8)	262k	51	$0.39	100.2	0.64	5.63	N/A
Qwen3 235B 2507 (Non-reasoning)	262k	51	$0.25	18.0	0.38	28.19	N/A
Qwen3 235B 2507 (Non-reasoning)	262k	51	$0.31	27.8	0.99	18.99	N/A
Qwen3 235B 2507 (Non-reasoning) (FP8)	131k	51	$0.40	54.0	0.53	9.80	N/A
Qwen3 235B 2507 (Non-reasoning) (FP8)	262k	51	$0.30	36.0	0.29	14.16	N/A
EXAONE 4.0 32B (Reasoning)	131k	51	$0.70	81.3	0.27	31.02	24.60
gpt-oss-20B (high) (Fast)	131k	49	$0.09	336.5	5.31	12.73	5.94
gpt-oss-20B (high)	131k	49	$0.13	260.7	13.38	22.97	7.67
gpt-oss-20B (high) Base	128k	49	$0.09	182.9	0.55	14.22	10.94
gpt-oss-20B (high) Vertex	131k	49	$0.13	328.5	0.15	7.76	6.09
gpt-oss-20B (high)	131k	49	$0.05	264.2	0.19	9.65	7.57
gpt-oss-20B (high)	131k	49	$0.09	352.0	0.58	7.68	5.68
gpt-oss-20B (high)	131k	49	$0.07	190.2	0.20	13.34	10.52
gpt-oss-20B (high)	131k	49	$0.09	191.1	0.53	13.61	10.47
gpt-oss-20B (high)	131k	49	$0.20	1,117.5	0.23	2.47	1.79
gpt-oss-20B (high)	131k	49	$0.09	233.8	7.28	17.98	8.55
gpt-oss-20B (high)	128k	49	$0.23	94.1	26.51	53.08	21.26
DeepSeek V3.1 (Non-reasoning)	128k	49	$0.48	19.2	2.87	28.93	N/A
DeepSeek V3.1 (Non-reasoning) (FP8)	164k	49	$0.47	53.9	0.35	9.63	N/A
DeepSeek V3.1 (Non-reasoning)	33k	49	$3.38	133.4	2.25	6.00	N/A
Kimi K2	131k	49	$2.13	42.2	0.51	12.36	N/A
Kimi K2	131k	49	$0.97	51.6	0.64	10.33	N/A
Kimi K2	131k	49	$1.07	77.2	0.54	7.02	N/A
Kimi K2	131k	49	$0.88	44.4	0.26	11.51	N/A
Kimi K2	131k	49	$1.00	57.2	1.23	9.96	N/A
Kimi K2	131k	49	$1.50	25.4	0.58	20.26	N/A
Kimi K2	131k	49	$1.50	346.2	0.23	1.67	N/A
Kimi K2	131k	49	$1.50	19.1	0.72	26.84	N/A
Kimi K2	131k	49	$1.07	72.6	0.19	7.07	N/A
Gemini 2.5 Flash (AI_Studio)	1m	47	$0.85	222.0	0.32	2.57	N/A
Gemini 2.5 Flash (Vertex)	1m	47	$0.85	233.7	0.32	2.46	N/A
Gemini 2.5 Flash-Lite (Reasoning) (AI Studio)	1m	47	$0.17	439.0	8.72	9.86	N/A
GPT-4.1	1m	47	$3.50	117.0	0.53	4.80	N/A
GPT-4.1	1m	47	$3.50	146.4	0.84	4.26	N/A
Claude 4 Opus	200k	47	$30.00	21.9	2.83	25.66	N/A
Claude 4 Opus Vertex	200k	47	$30.00	42.0	1.85	13.76	N/A
Claude 4 Opus	200k	47	$30.00	39.6	1.95	14.57	N/A
Llama Nemotron Ultra Reasoning Base	131k	46	$0.90	37.1	0.68	68.06	53.90
Claude 4 Sonnet	1m	46	$6.00	69.1	1.25	8.48	N/A
Claude 4 Sonnet Vertex	1m	46	$6.00	68.3	1.50	8.82	N/A
Claude 4 Sonnet	1m	46	$6.00	70.3	1.52	8.63	N/A
GPT-4.1 mini	1m	46	$0.70	63.5	0.47	8.35	N/A
GPT-4.1 mini	1m	46	$0.70	104.2	0.83	5.63	N/A
Qwen3 Coder 480B (FP8)	262k	45	$1.63	54.2	0.43	9.66	N/A
Qwen3 Coder 480B	131k	45	$2.00	1,490.0	0.26	0.60	N/A
Qwen3 Coder 480B (FP8)	262k	45	$2.00	65.2	3.48	11.15	N/A
Qwen3 Coder 480B	262k	45	$0.75	62.6	0.62	8.60	N/A
Qwen3 Coder 480B Vertex	262k	45	$1.75	70.7	0.38	7.45	N/A
Qwen3 Coder 480B	262k	45	$0.79	114.1	0.51	4.89	N/A
Qwen3 Coder 480B (FP8)	262k	45	$0.70	64.2	0.34	8.12	N/A
Qwen3 Coder 480B (Turbo, FP4)	262k	45	$0.53	70.0	0.22	7.37	N/A
Qwen3 Coder 480B	262k	45	$1.10	67.9	0.75	8.12	N/A
Qwen3 Coder 480B (FP8)	262k	45	$1.10	84.9	0.51	6.40	N/A
Qwen3 Coder 480B (FP8)	262k	45	$2.00	46.7	0.43	11.13	N/A
Qwen3 30B 2507 (Non-reasoning)	262k	44	$0.15	125.0	0.59	4.59	N/A
Qwen3 30B 2507 (Non-reasoning) (FP8)	262k	44	$0.50	165.1	0.49	3.52	N/A
GPT-5 (minimal)	400k	44	$3.44	129.5	1.06	4.92	N/A
Solar Pro 2 (Reasoning)	66k	43	$0.50	101.6	1.33	25.93	19.68
QwQ-32B	131k	42	$0.20	110.6	3.46	30.49	22.52
QwQ-32B Fast	131k	42	$0.75	80.7	0.54	37.61	30.88
QwQ-32B Base	131k	42	$0.23	50.4	0.60	59.89	49.39
QwQ-32B	131k	42	$0.09	35.5	0.53	84.89	70.25
QwQ-32B	131k	42	$0.75	47.3	0.40	63.59	52.63
QwQ-32B	131k	42	$1.20	86.5	0.27	34.86	28.81
Llama 4 Maverick (FP8)	1m	42	$0.28	131.4	0.23	4.03	N/A
Llama 4 Maverick (FP8)	1m	42	$0.35	125.2	0.37	4.36	N/A
Llama 4 Maverick	32k	42	$0.30	2,309.9	0.26	0.48	N/A
Llama 4 Maverick	128k	42	$0.42	273.8	0.62	2.44	N/A
Llama 4 Maverick Vertex	524k	42	$0.55	248.2	0.32	2.34	N/A
Llama 4 Maverick (FP8)	128k	42	$0.61	140.7	0.34	3.89	N/A
Llama 4 Maverick (Base)	1m	42	$0.39	26.4	2.23	21.17	N/A
Llama 4 Maverick (FP8)	1m	42	$0.26	60.1	0.30	8.62	N/A
Llama 4 Maverick (Turbo, FP8)	8k	42	$0.50	824.9	0.21	0.81	N/A
Llama 4 Maverick (FP8)	1m	42	$0.34	82.3	0.46	6.53	N/A
Llama 4 Maverick (FP8)	1m	42	$0.39	141.3	0.81	4.35	N/A
Llama 4 Maverick	131k	42	$0.30	501.1	0.16	1.16	N/A
Llama 4 Maverick	131k	42	$0.92	693.8	0.36	1.08	N/A
Llama 4 Maverick	1m	42	$0.41	78.5	0.21	6.58	N/A
DeepSeek R1 0528 Qwen3 8B	131k	39	$0.06	91.7	0.38	27.64	21.81
DeepSeek R1 0528 Qwen3 8B	128k	39	$0.07	79.6	0.84	32.23	25.12
Mistral Medium 3	131k	39	$0.80	47.9	0.42	10.86	N/A
Mistral Medium 3	128k	39	$0.80	47.9	0.43	10.87	N/A
Mistral Medium 3.1	131k	38	$0.80	42.8	0.40	12.08	N/A
Magistral Medium	41k	38	$2.75	146.3	0.39	17.48	13.67
EXAONE 4.0 32B	131k	37	$0.70	76.2	0.28	6.85	N/A
Magistral Small	40k	36	$0.75	181.6	0.35	14.12	11.01
Qwen3 Coder 30B	262k	36	$0.15	127.2	0.54	4.47	N/A
Qwen3 Coder 30B	262k	36	$0.26	180.7	0.48	3.24	N/A
Gemini 2.5 Flash-Lite (AI Studio)	1m	35	$0.17	314.1	0.21	1.80	N/A
Nova Premier	1m	35	$5.00	75.9	0.87	7.46	N/A
Solar Pro 2	66k	33	$0.50	106.1	1.30	6.01	N/A
Llama 4 Scout	1m	33	$0.14	88.4	0.18	5.83	N/A
Llama 4 Scout	32k	33	$0.70	2,068.4	0.27	0.52	N/A
Llama 4 Scout	128k	33	$0.29	137.3	0.54	4.18	N/A
Llama 4 Scout Vertex	1m	33	$0.36	129.4	0.51	4.38	N/A
Llama 4 Scout	10m	33	$0.11	113.3	0.53	4.94	N/A
Llama 4 Scout	128k	33	$0.34	103.3	0.32	5.16	N/A
Llama 4 Scout (Base)	10m	33	$0.26	137.5	0.53	4.17	N/A
Llama 4 Scout	328k	33	$0.14	49.8	0.30	10.34	N/A
Llama 4 Scout	131k	33	$0.20	63.1	0.83	8.76	N/A
Llama 4 Scout	1m	33	$0.18	112.6	0.41	4.85	N/A
Llama 4 Scout	131k	33	$0.17	435.6	0.20	1.35	N/A
Llama 4 Scout	1m	33	$0.28	93.0	0.19	5.57	N/A
Llama 4 Scout	131k	33	$0.41	109.2	0.71	5.29	N/A
Mistral Small 3.2	131k	32	$0.15	140.0	0.30	3.87	N/A
Mistral Small 3.2 (FP8)	128k	32	$0.06	27.6	0.35	18.47	N/A
Command A	256k	32	$4.38	150.7	0.21	3.53	N/A
Command A	256k	32	$4.38	38.3	0.64	13.69	N/A
GPT-4.1 nano	1m	32	$0.17	158.0	0.40	3.56	N/A
GPT-4.1 nano	1m	32	$0.17	169.2	0.64	3.60	N/A
Devstral Medium	131k	31	$0.80	104.0	0.39	5.20	N/A
Llama 3.3 70B (FP8)	128k	31	$0.17	38.2	0.23	13.33	N/A
Llama 3.3 70B (FP8)	131k	31	$0.28	86.4	0.41	6.20	N/A
Llama 3.3 70B	128k	31	$0.94	2,143.8	0.24	0.47	N/A
Llama 3.3 70B	131k	31	$0.40	27.0	2.98	21.51	N/A
Llama 3.3 70B	128k	31	$0.71	177.0	0.54	3.37	N/A
Llama 3.3 70B Fast	128k	31	$0.38	199.5	0.57	3.08	N/A
Llama 3.3 70B Base	128k	31	$0.20	30.5	0.73	17.13	N/A
Llama 3.3 70B Vertex	128k	31	$0.72	115.6	0.21	4.53	N/A
Llama 3.3 70B Snowflake	8k	31	$0.58	148.5	0.32	3.69	N/A
Llama 3.3 70B	128k	31	$0.40	206.8	0.24	2.66	N/A
Llama 3.3 70B	128k	31	$0.71	41.0	0.43	12.64	N/A
Llama 3.3 70B	131k	31	$0.90	147.0	0.42	3.83	N/A
Llama 3.3 70B (Turbo, FP8)	131k	31	$0.06	41.4	0.36	12.44	N/A
Llama 3.3 70B	131k	31	$0.27	23.4	0.58	21.93	N/A
Llama 3.3 70B	128k	31	$0.60	142.2	0.29	3.80	N/A
Llama 3.3 70B	128k	31	$0.60	140.2	0.29	3.86	N/A
Llama 3.3 70B	131k	31	$0.20	36.1	0.60	14.43	N/A
Llama 3.3 70B	131k	31	$0.64	380.9	0.24	1.56	N/A
Llama 3.3 70B	128k	31	$0.75	369.3	0.42	1.78	N/A
Llama 3.3 70B Turbo	131k	31	$0.88	89.1	0.36	5.97	N/A
Llama 3.3 70B	24k	31	$0.78	27.1	0.93	19.40	N/A
Llama 3.1 405B	128k	29	$9.50	15.9	1.04	32.51	N/A
Llama 3.1 405B	131k	29	$4.00	78.1	2.27	8.67	N/A
Llama 3.1 405B Standard	128k	29	$2.40	25.6	1.83	21.32	N/A
Llama 3.1 405B Latency Optimized	128k	29	$3.00	75.9	0.44	7.03	N/A
Llama 3.1 405B Base	128k	29	$1.50	27.7	0.74	18.77	N/A
Llama 3.1 405B Vertex	128k	29	$7.75	25.1	0.41	20.29	N/A
Llama 3.1 405B	128k	29	$8.00	26.3	0.47	19.50	N/A
Llama 3.1 405B	131k	29	$3.00	78.9	0.55	6.88	N/A
Llama 3.1 405B	16k	29	$6.25	145.8	0.60	4.03	N/A
Llama 3.1 405B	128k	29	$7.50	31.4	0.89	16.81	N/A
Llama 3.1 405B Turbo	131k	29	$3.50	74.4	0.52	7.24	N/A
Phi-4	16k	28	$0.15	96.9	0.56	5.72	N/A
Phi-4	16k	28	$0.22	36.2	0.42	14.24	N/A
Phi-4	16k	28	$0.09	37.3	0.25	13.64	N/A
Llama 3.1 Nemotron 70B	131k	26	$0.17	30.9	0.62	16.79	N/A
Gemma 3 27B	131k	25	$0.29	52.4	0.42	9.97	N/A
Gemma 3 27B (AI_Studio)	128k	25	$0.00	47.0	0.61	11.25	N/A
Gemma 3 27B	131k	25	$0.11	30.0	0.43	17.08	N/A
Jamba 1.7 Large	256k	24	$3.50	45.1	0.83	11.92	N/A
Gemma 3 12B	131k	24	$0.06	40.3	0.31	12.70	N/A
Gemma 3 12B	80k	24	$0.40	77.5	0.59	7.05	N/A
Gemma 3n E4B	33k	18	$0.03	72.9	0.32	7.18	N/A
Gemma 3 4B	131k	18	$0.03	57.7	0.58	9.24	N/A
Granite 3.3 8B	128k	18	$0.09	103.9	0.37	5.18	N/A
Llama 3.2 11B (Vision)	128k	17	$0.16	155.5	0.46	3.67	N/A
Llama 3.2 11B (Vision)	128k	17	$0.37	72.2	0.35	7.27	N/A
Llama 3.2 11B (Vision)	131k	17	$0.05	42.5	0.22	11.98	N/A
GPT-4o mini	128k		$0.26	58.3	0.50	9.08	N/A
GPT-4o mini	128k		$0.26	68.4	1.00	8.31	N/A
o3-mini	200k		$1.93	158.4	13.46	16.62	N/A
o3-mini	200k		$1.93	160.0	12.03	15.16	N/A
o3-mini (high)	200k		$1.93	152.7	41.55	44.82	N/A
o3-mini (high)	200k		$1.93	158.3	37.77	40.93	N/A
o3-pro	200k		$35.00	20.1	121.55	146.40	N/A
o3-pro	200k		$35.00	22.2	128.66	151.21	N/A
Llama 3.2 90B (Vision)	128k		$0.72	49.5	0.50	10.60	N/A
Llama 3.2 90B (Vision) Vertex	128k		$0.00	26.9	0.19	18.80	N/A
Llama 3.2 90B (Vision)	128k		$2.04	35.3	0.35	14.52	N/A
Llama 3.2 90B (Vision)	33k		$0.36	25.3	0.37	20.10	N/A
Gemma 3n E2B (AI Studio)	32k		$0.00	47.1	0.27	10.89	N/A
Claude 4.1 Opus Thinking	200k		$30.00	19.1	3.55	134.63	104.87
Claude 4.1 Opus Thinking Vertex	200k		$30.00	40.4	1.64	63.52	49.50
Claude 4.1 Opus Thinking	200k		$30.00	22.5	1.52	112.50	88.78
Claude 4.1 Opus	200k		$30.00	20.1	3.60	28.42	N/A
Claude 4.1 Opus Vertex	200k		$30.00	42.1	1.63	13.51	N/A
Claude 4.1 Opus	200k		$30.00	23.7	1.61	22.69	N/A
Ministral 8B	131k		$0.10	164.5	0.30	3.34	N/A
Ministral 3B	131k		$0.04	265.1	0.30	2.19	N/A
Codestral (Jan '25)	262k		$0.45	169.5	0.31	3.26	N/A
Codestral (Jan '25) Vertex	128k		$0.45	124.8	0.18	4.19	N/A
Devstral Small	131k		$0.15	141.6	0.34	3.87	N/A
Devstral Small	128k		$0.12	123.2	0.55	4.61	N/A
Devstral Small	128k		$0.12	81.4	0.53	6.68	N/A
Grok 3	131k		$6.00	35.0	1.14	15.44	N/A
Grok 3 Fast	131k		$10.00	67.7	0.73	8.12	N/A
Grok 3	16k		$6.00	59.2	0.53	8.98	N/A
Grok 3 mini Reasoning (low)	131k		$0.35	141.4	0.59	18.27	14.15
Grok 3 mini Reasoning (low) Fast	131k		$1.45	188.5	0.57	13.84	10.61
Phi-4 Multimodal	128k		$0.00	17.4	0.34	29.05	N/A
Aya Expanse 32B	128k		$0.75	100.2	0.17	5.16	N/A
Aya Expanse 8B	8k		$0.75	135.1	0.14	3.84	N/A
Jamba 1.7 Mini	258k		$0.25	137.0	0.72	4.37	N/A

Key definitions

Back to Navigation

Context window: Maximum number of combined input & output tokens. Output tokens commonly have a significantly lower limit (varied by model).

Output Speed: Tokens per second received while the model is generating tokens (ie. after first chunk has been received from the API for models which support streaming).

Latency (Time to First Token): Time to first token received, in seconds, after API request sent. For reasoning models which share reasoning tokens, this will be the first reasoning token. For models which do not support streaming, this represents time to receive the completion.

Price: Price per token, represented as USD per million Tokens. Price is a blend of Input & Output token prices (3:1 ratio).

Output Price: Price per token generated by the model (received from the API), represented as USD per million Tokens.

Input Price: Price per token included in the request/message sent to the API, represented as USD per million Tokens.

Time period: Metrics are 'live' and are based on the past 72 hours of measurements, measurements are taken 8 times a day for single requests and 2 times per day for parallel requests.