トークン
同一モデルにおけるトークンプロバイダー間のレイテンシと価格の差（サンプルスナップショット）。
Prices on this page are listed in USD per 1M tokens unless a column says otherwise.
Same-model comparison
·
GPT-4o mini — Same-model comparison
Fastest · Groq
↗
200ms
P50 latency
Fastest
Best price · OpenRouter
↗
$0.15 / $0.60
Input / 1M · Output / 1M
GPT-4o mini — Official site
Providers are sorted by a balanced score: lower latency and lower average of input/output price per 1M rank higher; official badges break ties.
- 1
- Input / 1M
- $0.15
- Output / 1M
- $0.60
- P50 latency
- 200ms
- 2
- Input / 1M
- $0.15
- Output / 1M
- $0.60
- P50 latency
- 380ms
- 3
- Input / 1M
- $0.16
- Output / 1M
- $0.62
- P50 latency
- 310ms
- 4
- Input / 1M
- $0.15
- Output / 1M
- $0.60
- P50 latency
- 420ms
- 5
- Input / 1M
- $0.15
- Output / 1M
- $0.60
- P50 latency
- 450ms
- 6
- Input / 1M
- $0.18
- Output / 1M
- $0.64
- P50 latency
- 340ms
Claude Sonnet 4.6 — Same-model comparison
Best price · OpenRouter
↗
$3.00 / $15.00
Input / 1M · Output / 1M
Claude Sonnet 4.6 — Official site
Providers are sorted by a balanced score: lower latency and lower average of input/output price per 1M rank higher; official badges break ties.
- 1
- Input / 1M
- $3.00
- Output / 1M
- $15.00
- P50 latency
- 420ms
- 2
- Input / 1M
- $3.00
- Output / 1M
- $15.00
- P50 latency
- 440ms
- 3
- Input / 1M
- $3.00
- Output / 1M
- $15.00
- P50 latency
- 470ms
- 4
- Input / 1M
- $3.00
- Output / 1M
- $15.00
- P50 latency
- 490ms
DeepSeek V3 — Same-model comparison
Fastest · Groq
↗
180ms
P50 latency
Fastest
Best price · DeepSeek
↗
$0.11 / $0.28
Input / 1M · Output / 1M
Best price
DeepSeek V3 — Official site
Providers are sorted by a balanced score: lower latency and lower average of input/output price per 1M rank higher; official badges break ties.
- 1
- Input / 1M
- $0.22
- Output / 1M
- $0.35
- P50 latency
- 180ms
- 2
- Input / 1M
- $0.11
- Output / 1M
- $0.28
- P50 latency
- 350ms
- 3
- Input / 1M
- $0.16
- Output / 1M
- $0.30
- P50 latency
- 280ms
- 4
- Input / 1M
- $0.17
- Output / 1M
- $0.30
- P50 latency
- 300ms
- 5
- Input / 1M
- $0.17
- Output / 1M
- $0.31
- P50 latency
- 320ms
- 6
- Input / 1M
- $0.18
- Output / 1M
- $0.32
- P50 latency
- 310ms
- 7
- Input / 1M
- $0.18
- Output / 1M
- $0.32
- P50 latency
- 330ms
- 8
- Input / 1M
- $0.23
- Output / 1M
- $0.37
- P50 latency
- 260ms
- 9
- Input / 1M
- $0.19
- Output / 1M
- $0.33
- P50 latency
- 340ms
- 10
- Input / 1M
- $0.15
- Output / 1M
- $0.29
- P50 latency
- 400ms
- 11
- Input / 1M
- $0.19
- Output / 1M
- $0.32
- P50 latency
- 350ms
- 12
- Input / 1M
- $0.24
- Output / 1M
- $0.38
- P50 latency
- 290ms
- 13
- Input / 1M
- $0.19
- Output / 1M
- $0.33
- P50 latency
- 370ms
- 14
- Input / 1M
- $0.20
- Output / 1M
- $0.34
- P50 latency
- 360ms
- 15
- Input / 1M
- $0.20
- Output / 1M
- $0.34
- P50 latency
- 380ms
- 16
- Input / 1M
- $0.20
- Output / 1M
- $0.35
- P50 latency
- 410ms
- 17
- Input / 1M
- $0.12
- Output / 1M
- $0.24
- P50 latency
- 580ms
- 18
- Input / 1M
- $0.21
- Output / 1M
- $0.36
- P50 latency
- 430ms
- 19
- Input / 1M
- $0.21
- Output / 1M
- $0.36
- P50 latency
- 450ms
- 20
- Input / 1M
- $0.15
- Output / 1M
- $0.75
- P50 latency
- 420ms
Gemini 2.0 Flash — Same-model comparison
Gemini 2.0 Flash — Official site
Providers are sorted by a balanced score: lower latency and lower average of input/output price per 1M rank higher; official badges break ties.
- 1
- Input / 1M
- $0.10
- Output / 1M
- $0.40
- P50 latency
- 340ms
- 2
- Input / 1M
- $0.10
- Output / 1M
- $0.40
- P50 latency
- 350ms
- 3
- Input / 1M
- $0.10
- Output / 1M
- $0.40
- P50 latency
- 380ms
- 4
- Input / 1M
- $0.11
- Output / 1M
- $0.42
- P50 latency
- 320ms
Llama 3.3 70B Instruct — Same-model comparison
Fastest · Groq
↗
190ms
P50 latency
Fastest
Best price · OpenRouter
↗
$0.10 / $0.32
Input / 1M · Output / 1M
Llama 3.3 70B Instruct — Official site
Providers are sorted by a balanced score: lower latency and lower average of input/output price per 1M rank higher; official badges break ties.
- 1
- Input / 1M
- $0.10
- Output / 1M
- $0.32
- P50 latency
- 360ms
- 2
- Input / 1M
- $0.13
- Output / 1M
- $0.38
- P50 latency
- 190ms
- 3
- Input / 1M
- $0.14
- Output / 1M
- $0.39
- P50 latency
- 300ms
- 4
- Input / 1M
- $0.14
- Output / 1M
- $0.40
- P50 latency
- 330ms
- 5
- Input / 1M
- $0.12
- Output / 1M
- $0.35
- P50 latency
- 560ms
- 6
- Input / 1M
- $0.15
- Output / 1M
- $0.41
- P50 latency
- 320ms
Claude 3.5 Haiku — Same-model comparison
Best price · Anthropic
↗
$0.80 / $4.00
Input / 1M · Output / 1M
Best price
Claude 3.5 Haiku — Official site
Providers are sorted by a balanced score: lower latency and lower average of input/output price per 1M rank higher; official badges break ties.
- 1
- Input / 1M
- $0.80
- Output / 1M
- $4.00
- P50 latency
- 400ms
- 2
- Input / 1M
- $0.80
- Output / 1M
- $4.00
- P50 latency
- 420ms
- 3
- Input / 1M
- $0.80
- Output / 1M
- $4.00
- P50 latency
- 460ms
- 4
- Input / 1M
- $0.80
- Output / 1M
- $4.00
- P50 latency
- 480ms
- 5
- Input / 1M
- $0.85
- Output / 1M
- $4.10
- P50 latency
- 400ms
