MODEL OBSERVATORY

Explore the network

Models, providers and token prices. In one place.

Providers

8 providers

Scroll horizontally to compare all columns

Illustrative providers and 24 hours of service metrics ending 17 Oct 2026. Starting prices in USDC per million tokens; availability is synthetic probe reachability, latency is median time to first token.
Models	Input from / 1M	Output from / 1M	Delivery	Availability	Median TTFT	Offers seen
2	$0.050	$0.120	99.43%4,364 / 4,389	96.53%139 / 144 probes	466 ms	2 / 2 seen
4	$0.055	$0.132	99.35%11,396 / 11,470	99.31%143 / 144 probes	579 ms	4 / 4 seen
3	$0.060	$0.144	99.24%8,973 / 9,042	98.61%142 / 144 probes	597 ms	3 / 3 seen
3	$0.130	$0.390	99.16%9,924 / 10,008	97.92%141 / 144 probes	557 ms	2 / 3 seen
2	$0.040	$0.100	99.07%6,900 / 6,965	97.22%140 / 144 probes	614 ms	2 / 2 seen
2	$0.044	$0.110	99.75%7,534 / 7,553	96.53%139 / 144 probes	640 ms	2 / 2 seen
3	$0.120	$0.350	99.65%12,777 / 12,822	99.31%143 / 144 probes	809 ms	3 / 3 seen
1	$0.132	$0.385	99.55%4,226 / 4,245	98.61%142 / 144 probes	667 ms	1 / 1 seen

Demo service · 24 hours ending 17 Oct 2026. Delivery = completed requests; availability = probe reachability, not uptime; latency = median first-token time. Offers seen is a separate catalog snapshot. No live service score is assigned.