GoCompare

models

click a column to sort · click a scatter dot to find its row

model	usage	in / out / cache $ per 1M	cost / request	req / 5h	req / mo	intel	coding	agentic	tok/s
Grok 4.5 xAI	$15	in$8.00out$24.00cache$1.20	$0.0999×4	120	600	55.8	72.4	48.9	58
GPT 5.6 Luna OpenAI · ≤ 272K tokens	$15	in$0.800out$4.80cache$0.080cw$1.00	$0.00586×4	2,050	10,250	52.3	71.4	46.9	153
GPT 5.6 Luna OpenAI · > 272K tokens	$15	in$1.60out$7.20cache$0.160cw$2.00	$0.0112×4	2,050	10,250	52.3	71.4	46.9	153
GLM-5.2 Zhipu	$60	in$1.40out$4.40cache$0.260	$0.0152	880	4,300	52.6	68.8	45.7	115
GLM-5.1 Zhipu	$60	in$1.40out$4.40cache$0.260	$0.0152	880	4,300	41.0	55.8	30.6	67
Kimi K3 Moonshot AI	$15	in$12.00out$60.00cache$1.20	$0.1224×4	110	490	59.7	76.2	54.3	40
Kimi K2.7 Code Moonshot AI	$60	in$0.950out$4.00cache$0.190	$0.0121	1,350	6,750	43.0	60.8	30.3	37
Kimi K2.6 Moonshot AI	$60	in$0.950out$4.00cache$0.160	$0.0104	1,150	5,750	45.1	61.8	31.2	49
MiMo-V2.5 Xiaomi	$60	in$0.140out$0.280cache$0.0028	$0.000399	30,100	150,400	38.0	56.8	24.4	89
MiMo-V2.5-Pro Xiaomi	$15	in$1.74out$3.48cache$0.015	$0.00368×4	3,250	16,300	42.9	60.2	29.5	46
MiniMax M3 MiniMax	$60	in$0.300out$1.20cache$0.060	$0.00374	3,200	16,000	45.4	58.6	36.1	80
MiniMax M2.7 MiniMax	$60	in$0.300out$1.20cache$0.060cw$0.375	$0.00354	3,400	17,000	38.9	52.6	25.9	64
Qwen3.8 Max Alibaba	$15	in$8.00out$24.00cache$1.00cw$10.00	$0.0742×4	160	810	58.1	71.8	58.4	47
Qwen3.7 Max Alibaba	$60	in$2.50out$7.50cache$0.500cw$3.13	$0.0355	340	1,690	46.7	66.0	30.9	204
Qwen3.7 Plus Alibaba · ≤ 256K tokens	$60	in$0.400out$1.60cache$0.040cw$0.500	$0.00278	4,300	21,600	39.4	55.9	20.7	57
Qwen3.7 Plus Alibaba · > 256K tokens	$60	in$1.20out$4.80cache$0.120cw$1.50	$0.00835	4,300	21,600	39.4	55.9	20.7	57
Qwen3.6 Plus Alibaba · ≤ 256K tokens	$60	in$0.500out$3.00cache$0.050cw$0.625	$0.00367	3,300	16,300	40.5	54.5	29.0	56
Qwen3.6 Plus Alibaba · > 256K tokens	$60	in$2.00out$6.00cache$0.200cw$2.50	$0.0135	3,300	16,300	40.5	54.5	29.0	56
DeepSeek V4 Pro DeepSeek	$15	in$1.74out$3.48cache$0.015	$0.00350×4	3,450	17,150	53.2	68.8	49.6	80
DeepSeek V4 Flash DeepSeek	$60	in$0.140out$0.280cache$0.0028	$0.000379	31,650	158,150	51.8	69.1	48.4	120
Hy3 Tencent	$60	in$0.140out$0.580cache$0.035	$0.00279	4,300	21,500	42.2	58.8	31.4	64

graph

pick the axes · defaults to intelligence vs effective cost · click a dot to find its row

x axislog scale

y axislog scale

$60 tier$15 tierxAIOpenAIZhipuMoonshot AIXiaomiMiniMaxAlibabaDeepSeekTencent

calculator

typical patterns pre-filled · adjust tokens to your workload

model

input tokens

cached tokens

output tokens

effective cost / request—

listed cost / request—

effective cost, 100 requests—

requests per $12 window (derived)—

requests per month (derived)—

derived window counts bill at the listed rate against the usage limits ($12 / 5h, $30 / week, $15 or $60 / month). The published opencode estimates are observed numbers and differ from these.

benchmarks

artificial analysis composite indices

intelligence index18

Kimi K359.7

Qwen3.8 Max58.1

Grok 4.555.8

DeepSeek V4 Pro53.2

GLM-5.252.6

GPT 5.6 Luna52.3

DeepSeek V4 Flash51.8

Qwen3.7 Max46.7

MiniMax M345.4

Kimi K2.645.1

Kimi K2.7 Code43.0

MiMo-V2.5-Pro42.9

Hy342.2

GLM-5.141.0

Qwen3.6 Plus40.5

Qwen3.7 Plus39.4

MiniMax M2.738.9

MiMo-V2.538.0

coding index18

Kimi K376.2

Grok 4.572.4

Qwen3.8 Max71.8

GPT 5.6 Luna71.4

DeepSeek V4 Flash69.1

GLM-5.268.8

DeepSeek V4 Pro68.8

Qwen3.7 Max66.0

Kimi K2.661.8

Kimi K2.7 Code60.8

MiMo-V2.5-Pro60.2

Hy358.8

MiniMax M358.6

MiMo-V2.556.8

Qwen3.7 Plus55.9

GLM-5.155.8

Qwen3.6 Plus54.5

MiniMax M2.752.6

agentic index18

Qwen3.8 Max58.4

Kimi K354.3

DeepSeek V4 Pro49.6

Grok 4.548.9

DeepSeek V4 Flash48.4

GPT 5.6 Luna46.9

GLM-5.245.7

MiniMax M336.1

Hy331.4

Kimi K2.631.2

Qwen3.7 Max30.9

GLM-5.130.6

Kimi K2.7 Code30.3

MiMo-V2.5-Pro29.5

Qwen3.6 Plus29.0

MiniMax M2.725.9

MiMo-V2.524.4

Qwen3.7 Plus20.7