🛡️

Provider Trust

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
burncloud
burncloud
484
484391API-measured
2
requesty
requesty
393
393315API-measured
3
Fireworks AI
hf · official · openrouter
259
27220298.8%919msAPI-measured
4
Amazon Bedrock
official · openrouter
224
23822099.6%634msAPI-measured
5
Azure
official · openrouter
143
16896100%1200msAPI-measured
6
Novita
hf · novita-ai · official · openrouter
138
2995999.9%985msFP4API-measured
7
Google
official · openrouter
123
16010699.8%727msAPI-measured
8
DeepInfra
hf · official · openrouter
102
1495199.8%645msFP4API-measured
9
OpenAI
official · openrouter
78
896999.6%2702msAPI-measured
10
Mistral AI
official
51
5146Official list
11
Together AI
hf · official · openrouter
51
621597.8%535msFP8API-measured
12
xAI
official
40
4038Official list
13
Replicate
official
40
4021Official list
14
Oracle Cloud
official
35
3535Official list
15
Cloudflare
official · openrouter
30
348100.0%913msFP8API-measured
16
Nebius
official · openrouter
29
321198.4%488msFP4API-measured
17
IBM watsonx
official
28
2824Official list
18
Alibaba
official · openrouter
28
291299.9%1030msFP8API-measured
19
Databricks
official
25
2525Official list
20
GMI Cloud
official · openrouter
25
261499.2%2252msFP8API-measured
21
Anthropic
official
24
2421Official list
22
Moonshot
official · openrouter
23
2619100%1755msINT4API-measured
23
Nscale
hf · official
23
3112606msAPI-measured
24
Lambda
official
20
2019Official list
25
20
2012API-measured