About
The latest GPT-4 Turbo model with vision capabilities. Vision requests can now use JSON mode and function calling. Training data: up to December 2023.
Related Models
Pricing
Input
$10.00
per 1M tokens
Output
$30.00
per 1M tokens
Blended
$15.00
per 1M tokens
Cheaper than 4% of models. Median price is $0.54/1M tokens.
Cost Calculator
Tokens per day1M
100K100M
Daily
$15.00
Monthly
$450.00
vs. Similar Models
GPT-5 nano (minimal)Q:+0.1
$0.14-99%
Olmo 3.1 32B ThinkQ:+0.2
$0.24-98%
Pixtral LargeQ:+0.2
$3.00-80%
Cohere: Command AQ:-0.2
$4.38-71%
Performance
38
tokens/sec
Faster than 7% of models
1.81
seconds
Faster than 25% of models
1.81
seconds
Faster than 52% of models
Market Median
94 tok/s
59% slower
Median TTFT
1.11s
62% slower
Throughput/Dollar
3
tok/s per $/1M
Speed Comparison
Qwen3.6 Max Preview
38 tok/s-1%
Magistral Medium 1.2
39 tok/s+1%
Seed-OSS-36B-Instruct
38 tok/s-1%
Context Window
128K
tokens
Larger than 16% of models
Max Output
4K
tokens
3% of context
Benchmarks
MMLU-Pro
69.4%
GPQA DiamondNot evaluated
HLE
3.3%
LiveCodeBench
29.1%
SciCode
31.9%
TerminalBench HardNot evaluated
MATH-500
73.7%
AIME
15.0%
AIME 2025Not evaluated
IFBenchNot evaluated
Long Context RecallNot evaluated
Tau2Not evaluated
Market AverageTop Score