GPT Models Pricing

tags :

Comparison between various flag ship models #

ref

Price Comparison #

Quality #

GPT-4 Turbo is of higher quality compared to average, with a MMLU score of 0.864 and a Quality Index across evaluations of 94.

Price #

GPT-4 Turbo is more expensive compared to average with a price of $15.00 per 1M Tokens (blended 3:1). GPT-4 Turbo Input token price: $10.00, Output token price: $30.00 per 1M Tokens.

Speed: #

GPT-4 Turbo is slower compared to average, with a throughput of 20.2 tokens per second.

Latency: #

GPT-4 Turbo has a higher latency compared to average, taking 0.57s to receive the first token (TTFT).

Context Window: #

GPT-4 Turbo has a larger context windows than average, with a context window of 130k tokens.

OCR of Images #

2024-05-16_10-21-59_screenshot.png #

QUALITY SPEED PRICE Higher is better Throughput in Tokens per Second; Higher is better USD per 1M Tokens; Lower is better 30 100 94 94 88 88 156 122 13 96 75 62 15 10.5 12 7.5 81 76 75 72 65 65 65 8 9 30 29 0.2 0.5 0.5 0.8 0.8 0.9 9 9 Al G X U R  S - Al 00 M 00 Al 9 00 M Al 9 00 M Al  b 00 G M 9 Al e o d  S S 8  e 2 a 2 E 8 8 8 d 8 df  $ S e e " R  o - ) 1 8 / 8 d ) 2 A 9 > e S e X e 8 A e 3 s