🛡️

Provider Trust

i

Which providers are reliable — not which models are strong.

Official list API-measured TKX probe · soon Self-reported
Provider Trust
#Provider Models Prices Cheapest on Uptime Latency Precision Evidence
1
requesty
requesty
442
442351API-measured
2
burncloud
burncloud
387
387281API-measured
3
Fireworks AI
hf · official · openrouter
276
28921399.8%785msAPI-measured
4
Amazon Bedrock
official · openrouter
255
285248100%894msAPI-measured
5
Azure
official · openrouter
186
20811499.9%1451msAPI-measured
6
DeepInfra
hf · official · openrouter
147
2357099.8%587msFP4API-measured
7
Novita
hf · novita-ai · official · openrouter
143
3335899.9%1088msFP4API-measured
8
Google
official · openrouter
130
17310499.9%1454msAPI-measured
9
122
12238API-measured
10
Together AI
hf · official · openrouter
95
1182799.8%575msAPI-measured
11
OpenAI
official · openrouter
86
8861100.0%549msAPI-measured
12
Mistral AI
official · openrouter
72
7261100%952msNVFP4API-measured
13
aihubmix
official
72
7234Official list
14
Databricks
official
60
6060Official list
15
Nebius
official · openrouter
55
571880.4%822msFP4API-measured
16
xAI
official
40
4030Official list
17
Replicate
official
40
4016Official list
18
WandB
official
36
367Official list
19
Oracle Cloud
official
35
3535Official list
20
Alibaba
official · openrouter
34
4311100%1103msFP8API-measured
21
Cloudflare
official · openrouter
32
35999.9%1114msFP8API-measured
22
GMI Cloud
official · openrouter
31
328100.0%2439msFP8API-measured
23
IBM watsonx
official
30
3025Official list
24
Anthropic
official · openrouter
28
3225100.0%3087msAPI-measured
25
27
2711Official list