Inference providers

204 providers across OpenRouter, Vercel, Requesty and Hugging Face, ranked by the tokens they carried in the last week. "Price rank" is where the provider's list price sits among every offer of the same model on the same gateway: how many models it is the cheapest on, and its median position. "Off card" is how far its list price is from the price the model's own author charges, the number discount ads quote.

#1 Stealth
stealth · openrouter.ai
tokens / week
4.476T · 29.6%
models · endpoints
3 · 3
gateways
openrouter vercel
price rank
–
off card, median
–
cache hit
84.0%
uptime · p50 · speed
100.0% · 6785ms · 20t/s
data policy
trains on some · retains on some
#2 Tencent CN
tencent · tencentcloud.com
tokens / week
1.348T · 8.9%
models · endpoints
24 · 35
gateways
openrouter requesty vercel
price rank
cheapest on 7 of 22 · median cheaper than 74%
off card, median
–
cache hit
94.4%
uptime · p50 · speed
100.0% · 2018ms · 48t/s
data policy
trains on some · retains on some
#3 DeepInfra US
deepinfra · deepinfra.com
tokens / week
924.57B · 6.1%
models · endpoints
285 · 420
gateways
hf openrouter requesty vercel
price rank
cheapest on 218 of 336 · median cheapest
off card, median
20% off over 48
cache hit
88.3%
uptime · p50 · speed
98.5% · 596ms · 44t/s
data policy
no training · retains on some
#4 Xiaomi CN
xiaomi · xiaomimimo.com
tokens / week
688.84B · 4.6%
models · endpoints
5 · 15
gateways
openrouter requesty vercel
price rank
cheapest on 3 of 12 · median cheaper than 60%
off card, median
–
cache hit
92.6%
uptime · p50 · speed
99.5% · 4443ms · 42t/s
data policy
trains on some · retains on some
#5 DeepSeek CN
deepseek · deepseek.com
tokens / week
555.48B · 3.7%
models · endpoints
11 · 25
gateways
openrouter requesty vercel
price rank
cheapest on 2 of 22 · median cheaper than 57%
off card, median
–
cache hit
89.8%
uptime · p50 · speed
99.8% · 1204ms · 93t/s
data policy
trains on some · retains on some
#6 Together US
together · together.ai
tokens / week
545.42B · 3.6%
models · endpoints
68 · 75
gateways
hf openrouter requesty
price rank
cheapest on 7 of 62 · median cheaper than 50%
off card, median
card over 14
cache hit
89.1%
uptime · p50 · speed
95.5% · 390ms · 83t/s
data policy
no training · no retention
#7 OpenAI US
openai · openai.com
tokens / week
516.57B · 3.4%
models · endpoints
144 · 291
gateways
openrouter requesty vercel
price rank
cheapest on 111 of 176 · median cheapest
off card, median
–
cache hit
65.9%
uptime · p50 · speed
98.9% · 1943ms · 68t/s
data policy
no training · retains on some
#8 Wafer US
wafer · wafer.ai
tokens / week
417.31B · 2.8%
models · endpoints
17 · 38
gateways
openrouter vercel
price rank
cheapest on 5 of 38 · median cheaper than 91%
off card, median
72% off over 26
cache hit
84.1%
uptime · p50 · speed
99.8% · 540ms · 67t/s
data policy
no training · retains on some
#9 InferenceNet US
inference-net · inference.net
tokens / week
388.09B · 2.6%
models · endpoints
12 · 38
gateways
openrouter vercel
price rank
cheapest on 11 of 38 · median cheaper than 83%
off card, median
31% off over 19
cache hit
91.1%
uptime · p50 · speed
99.7% · 449ms · 96t/s
data policy
trains on some · retains on some
#10 StepFun
stepfun · stepfun.ai
tokens / week
367.98B · 2.4%
models · endpoints
3 · 5
gateways
openrouter vercel
price rank
cheapest on 2 of 4 · median cheaper than 65%
off card, median
55% off over 1
cache hit
94.6%
uptime · p50 · speed
99.3% · 1744ms · 145t/s
data policy
trains on some · retains on some
#11 Relace
relace · relace.ai
tokens / week
361.59B · 2.4%
models · endpoints
15 · 21
gateways
openrouter vercel
price rank
cheapest on 10 of 19 · median cheapest
off card, median
85% off over 12
cache hit
90.0%
uptime · p50 · speed
99.7% · 814ms · 84t/s
data policy
no training · no retention
#12 Google US
google-vertex · google.com
tokens / week
336.42B · 2.2%
models · endpoints
63 · 130
gateways
openrouter
price rank
cheapest on 45 of 124 · median cheaper than 60%
off card, median
+10% over 42
cache hit
56.5%
uptime · p50 · speed
92.6% · 1674ms · 72t/s
data policy
no training · retains on some
#13 Novita US
novita · novita.ai
tokens / week
290.38B · 1.9%
models · endpoints
212 · 273
gateways
hf openrouter requesty vercel
price rank
cheapest on 53 of 204 · median cheaper than 55%
off card, median
card over 33
cache hit
81.0%
uptime · p50 · speed
98.1% · 1271ms · 49t/s
data policy
no training · no retention
#14 TypeSafe
typesafe · typesafe.ai
tokens / week
210.69B · 1.4%
models · endpoints
3 · 3
gateways
openrouter requesty
price rank
–
off card, median
–
cache hit
0.0%
uptime · p50 · speed
100.0% · 121ms · 0t/s
data policy
no training · retains on some
#15 CoreWeave US
coreweave · coreweave.com
tokens / week
200.26B · 1.3%
models · endpoints
30 · 32
gateways
openrouter
price rank
cheapest on 3 of 31 · median cheaper than 67%
off card, median
23% off over 13
cache hit
80.6%
uptime · p50 · speed
98.9% · 386ms · 103t/s
data policy
no training · no retention
#16 Claude Platform on AWS US
claude-on-aws · anthropic.com
tokens / week
192.27B · 1.3%
models · endpoints
12 · 12
gateways
openrouter
price rank
cheapest on 12 of 12 · median cheapest
off card, median
card over 12
cache hit
77.3%
uptime · p50 · speed
100.0% · 1576ms · 57t/s
data policy
no training · retains on some
#17 DigitalOcean US
digitalocean · digitalocean.com
tokens / week
189.75B · 1.3%
models · endpoints
30 · 38
gateways
openrouter vercel
price rank
cheapest on 3 of 38 · median cheaper than 63%
off card, median
7% off over 18
cache hit
88.3%
uptime · p50 · speed
99.5% · 865ms · 66t/s
data policy
no training · retains on some
#18 Fireworks US
fireworks · fireworks.ai
tokens / week
188.48B · 1.2%
models · endpoints
44 · 91
gateways
openrouter requesty vercel
price rank
cheapest on 15 of 80 · median cheaper than 57%
off card, median
card over 26
cache hit
86.6%
uptime · p50 · speed
98.3% · 983ms · 102t/s
data policy
no training · no retention
#19 OpenInference US
open-inference · openinference.ai
tokens / week
184.85B · 1.2%
models · endpoints
6 · 7
gateways
openrouter
price rank
cheapest on 4 of 7 · median cheapest
off card, median
77% off over 6
cache hit
84.8%
uptime · p50 · speed
99.5% · 1108ms · 24t/s
data policy
no training · no retention
#20 Alibaba SG
alibaba · alibabacloud.com
tokens / week
179.3B · 1.2%
models · endpoints
118 · 145
gateways
openrouter requesty vercel
price rank
cheapest on 28 of 83 · median cheaper than 50%
off card, median
card over 18
cache hit
77.4%
uptime · p50 · speed
99.5% · 1170ms · 57t/s
data policy
no training · retains on some
#21 Minimax SG
minimax · minimax.io
tokens / week
153.71B · 1.0%
models · endpoints
21 · 33
gateways
openrouter vercel
price rank
cheapest on 7 of 18 · median cheaper than 75%
off card, median
–
cache hit
92.0%
uptime · p50 · speed
99.4% · 1185ms · 47t/s
data policy
no training · retains on some
#22 Amazon Bedrock US
amazon-bedrock · amazon.com
tokens / week
148.54B · 1.0%
models · endpoints
47 · 84
gateways
openrouter
price rank
cheapest on 31 of 78 · median cheaper than 60%
off card, median
+10% over 60
cache hit
80.2%
uptime · p50 · speed
98.2% · 1263ms · 75t/s
data policy
no training · retains on some
#23 Anthropic US
anthropic · anthropic.com
tokens / week
146.94B · 1.0%
models · endpoints
31 · 64
gateways
openrouter requesty vercel
price rank
cheapest on 52 of 56 · median cheapest
off card, median
–
cache hit
81.8%
uptime · p50 · speed
99.2% · 1252ms · 79t/s
data policy
no training · retains on some
#24 Azure US
azure · microsoft.com
tokens / week
137.7B · 0.9%
models · endpoints
82 · 177
gateways
openrouter requesty vercel
price rank
cheapest on 68 of 146 · median cheaper than 86%
off card, median
card over 103
cache hit
77.2%
uptime · p50 · speed
98.3% · 1702ms · 66t/s
data policy
no training · retains on some
#25 Morph US
morph · morphllm.com
tokens / week
135.03B · 0.9%
models · endpoints
19 · 33
gateways
openrouter vercel
price rank
cheapest on 4 of 29 · median cheaper than 67%
off card, median
25% off over 14
cache hit
85.9%
uptime · p50 · speed
98.0% · 966ms · 48t/s
data policy
no training · no retention
#26 Meta US
meta · facebook.com
tokens / week
134.24B · 0.9%
models · endpoints
11 · 19
gateways
openrouter requesty vercel
price rank
cheapest on 3 of 3 · median cheapest
off card, median
–
cache hit
89.0%
uptime · p50 · speed
99.3% · 2086ms · 111t/s
data policy
trains on some · retains on some
#27 PrimeIntellect
primeintellect · primeintellect.ai
tokens / week
119.27B · 0.8%
models · endpoints
26 · 26
gateways
openrouter
price rank
cheapest on 8 of 21 · median cheaper than 92%
off card, median
65% off over 7
cache hit
82.2%
uptime · p50 · speed
99.2% · 1018ms · 59t/s
data policy
no training · no retention
#28 Parasail US
parasail · parasail.io
tokens / week
109.34B · 0.7%
models · endpoints
62 · 89
gateways
openrouter requesty vercel
price rank
cheapest on 16 of 82 · median cheaper than 57%
off card, median
card over 27
cache hit
75.8%
uptime · p50 · speed
99.2% · 636ms · 82t/s
data policy
no training · no retention
#29 Cohere US
cohere · cohere.com
tokens / week
104.49B · 0.7%
models · endpoints
40 · 41
gateways
hf openrouter vercel
price rank
cheapest on 1 of 6 · median cheaper than 61%
off card, median
card over 5
cache hit
75.3%
uptime · p50 · speed
99.9% · 283ms · 77t/s
data policy
no training · retains on some
#30 Crusoe US
crusoe · crusoe.ai
tokens / week
90.24B · 0.6%
models · endpoints
17 · 19
gateways
openrouter vercel
price rank
cheapest on 1 of 19 · median cheaper than 50%
off card, median
card over 7
cache hit
84.9%
uptime · p50 · speed
99.2% · 497ms · 90t/s
data policy
trains on some · retains on some
#31 ModelRun US
modelrun · modular.com
tokens / week
90.06B · 0.6%
models · endpoints
7 · 8
gateways
openrouter
price rank
cheapest on 0 of 8 · median cheaper than 46%
off card, median
card over 6
cache hit
85.4%
uptime · p50 · speed
97.2% · 557ms · 140t/s
data policy
no training · no retention
#32 Google AI Studio US
google-ai-studio · google.com
tokens / week
89.64B · 0.6%
models · endpoints
32 · 63
gateways
openrouter
price rank
cheapest on 23 of 54 · median cheaper than 71%
off card, median
–
cache hit
56.4%
uptime · p50 · speed
96.3% · 2666ms · 77t/s
data policy
no training · retains on some
#33 Particle AI US
particle-ai · particle.ai
tokens / week
81.63B · 0.5%
models · endpoints
3 · 4
gateways
openrouter
price rank
cheapest on 0 of 3 · median cheaper than 90%
off card, median
83% off over 3
cache hit
81.7%
uptime · p50 · speed
100.0% · 746ms · 119t/s
data policy
no training · no retention
#34 StreamLake CN
streamlake · streamlake.ai
tokens / week
80.12B · 0.5%
models · endpoints
36 · 44
gateways
openrouter vercel
price rank
cheapest on 4 of 40 · median cheaper than 67%
off card, median
25% off over 23
cache hit
82.1%
uptime · p50 · speed
98.8% · 1748ms · 43t/s
data policy
trains on some · retains on some
#35 Z.AI SG
z-ai · z.ai
tokens / week
79.51B · 0.5%
models · endpoints
15 · 15
gateways
openrouter
price rank
cheapest on 2 of 10 · median cheaper than 44%
off card, median
–
cache hit
79.8%
uptime · p50 · speed
99.9% · 1975ms · 41t/s
data policy
no training · no retention
#36 Modal US
modal · modal.com
tokens / week
73.81B · 0.5%
models · endpoints
14 · 20
gateways
openrouter vercel
price rank
cheapest on 5 of 20 · median cheaper than 45%
off card, median
card over 7
cache hit
88.6%
uptime · p50 · speed
99.8% · 546ms · 112t/s
data policy
no training · no retention
#37 Decart US
decart · decart.ai
tokens / week
70.87B · 0.5%
models · endpoints
6 · 7
gateways
openrouter
price rank
cheapest on 0 of 7 · median cheaper than 75%
off card, median
51% off over 7
cache hit
86.2%
uptime · p50 · speed
97.9% · 1134ms · 90t/s
data policy
no training · no retention
#38 Mistral FR
mistral · mistral.ai
tokens / week
67.65B · 0.4%
models · endpoints
62 · 112
gateways
openrouter requesty vercel
price rank
cheapest on 40 of 81 · median cheaper than 67%
off card, median
card over 9
cache hit
84.0%
uptime · p50 · speed
98.5% · 497ms · 92t/s
data policy
no training · retains on some
#39 Upstage
upstage · upstage.ai
tokens / week
66.85B · 0.4%
models · endpoints
5 · 9
gateways
openrouter
price rank
cheapest on 8 of 8 · median cheapest
off card, median
–
cache hit
68.3%
uptime · p50 · speed
100.0% · 1675ms · 13t/s
data policy
no training · retains on some
#40 Ionstream US
ionstream · ionstream.ai
tokens / week
48.1B · 0.3%
models · endpoints
5 · 7
gateways
openrouter
price rank
cheapest on 0 of 7 · median cheaper than 74%
off card, median
27% off over 6
cache hit
89.0%
uptime · p50 · speed
99.5% · 967ms · 57t/s
data policy
no training · no retention
#41 BaseTen US
baseten · baseten.co
tokens / week
46.73B · 0.3%
models · endpoints
49 · 91
gateways
hf openrouter vercel
price rank
cheapest on 9 of 75 · median cheaper than 50%
off card, median
card over 23
cache hit
74.7%
uptime · p50 · speed
99.3% · 456ms · 118t/s
data policy
no training · no retention
#42 Cloudflare US
cloudflare · cloudflare.com
tokens / week
44.34B · 0.3%
models · endpoints
24 · 27
gateways
openrouter requesty
price rank
cheapest on 2 of 21 · median cheaper than 50%
off card, median
card over 11
cache hit
84.9%
uptime · p50 · speed
99.6% · 637ms · 43t/s
data policy
no training · retains on some
#43 Perplexity US
perplexity · perplexity.ai
tokens / week
41.21B · 0.3%
models · endpoints
10 · 20
gateways
openrouter requesty vercel
price rank
cheapest on 0 of 1 · median cheaper than 0%
off card, median
–
cache hit
0.0%
uptime · p50 · speed
100.0% · 2765ms · 44t/s
data policy
no training · no retention
#44 Makora
makora · makora.com
tokens / week
37.68B · 0.2%
models · endpoints
13 · 17
gateways
openrouter
price rank
cheapest on 3 of 17 · median cheaper than 64%
off card, median
7% off over 14
cache hit
69.2%
uptime · p50 · speed
97.3% · 832ms · 66t/s
data policy
no training · no retention
#45 GMICloud US
gmicloud · gmicloud.ai
tokens / week
37.28B · 0.2%
models · endpoints
39 · 64
gateways
openrouter vercel
price rank
cheapest on 23 of 61 · median cheaper than 77%
off card, median
20% off over 25
cache hit
79.8%
uptime · p50 · speed
92.3% · 2048ms · 50t/s
data policy
trains on some · retains on some
#46 xAI US
xai · x.ai
tokens / week
29.53B · 0.2%
models · endpoints
68 · 95
gateways
openrouter requesty vercel
price rank
cheapest on 23 of 39 · median cheapest
off card, median
–
cache hit
84.3%
uptime · p50 · speed
97.1% · 2151ms · 59t/s
data policy
no training · retains on some
#47 SiliconFlow SG
siliconflow · siliconflow.com
tokens / week
26.92B · 0.2%
models · endpoints
45 · 45
gateways
openrouter
price rank
cheapest on 8 of 43 · median cheaper than 64%
off card, median
5% off over 19
cache hit
77.2%
uptime · p50 · speed
98.3% · 1569ms · 31t/s
data policy
no training · no retention
#48 Sail Research US
sail-research · sailresearch.com
tokens / week
25.0B · 0.2%
models · endpoints
13 · 23
gateways
openrouter vercel
price rank
cheapest on 0 of 23 · median cheaper than 63%
off card, median
46% off over 18
cache hit
77.8%
uptime · p50 · speed
99.1% · 964ms · 55t/s
data policy
trains on some · retains on some
#49 Phala US
phala · redpill.ai
tokens / week
24.96B · 0.2%
models · endpoints
28 · 31
gateways
openrouter
price rank
cheapest on 3 of 31 · median cheaper than 52%
off card, median
10% off over 17
cache hit
80.0%
uptime · p50 · speed
98.7% · 1430ms · 56t/s
data policy
no training · no retention
#50 Nebius NL
nebius · nebius.com
tokens / week
22.94B · 0.2%
models · endpoints
46 · 64
gateways
openrouter requesty vercel
price rank
cheapest on 13 of 55 · median cheaper than 59%
off card, median
card over 10
cache hit
71.2%
uptime · p50 · speed
93.2% · 882ms · 82t/s
data policy
no training · no retention
#51 Venice US
venice · venice.ai
tokens / week
22.18B · 0.1%
models · endpoints
42 · 44
gateways
openrouter
price rank
cheapest on 3 of 43 · median cheaper than 50%
off card, median
card over 23
cache hit
66.9%
uptime · p50 · speed
98.6% · 938ms · 52t/s
data policy
no training · no retention
#52 Inceptron SE
inceptron · inceptron.io
tokens / week
17.13B · 0.1%
models · endpoints
11 · 17
gateways
openrouter requesty vercel
price rank
cheapest on 3 of 17 · median cheaper than 68%
off card, median
27% off over 7
cache hit
85.9%
uptime · p50 · speed
99.9% · 689ms · 70t/s
data policy
no training · retains on some
#53 Moonshot AI SG
moonshotai · moonshot.ai
tokens / week
13.39B · 0.1%
models · endpoints
7 · 17
gateways
openrouter vercel
price rank
cheapest on 1 of 16 · median cheaper than 43%
off card, median
–
cache hit
92.3%
uptime · p50 · speed
94.0% · 3180ms · 37t/s
data policy
no training · no retention
#54 Friendli US
friendli · friendli.ai
tokens / week
11.73B · 0.1%
models · endpoints
10 · 11
gateways
openrouter vercel
price rank
cheapest on 1 of 11 · median cheaper than 49%
off card, median
card over 5
cache hit
64.4%
uptime · p50 · speed
99.9% · 399ms · 87t/s
data policy
trains on some · retains on some
#55 NextBit ES
nextbit · nextbit256.com
tokens / week
11.59B · 0.1%
models · endpoints
15 · 16
gateways
openrouter
price rank
cheapest on 4 of 14 · median cheaper than 50%
off card, median
15% off over 6
cache hit
47.3%
uptime · p50 · speed
99.2% · 1156ms · 51t/s
data policy
no training · no retention
#56 Baidu CN
baidu · baidu.com
tokens / week
11.15B · 0.1%
models · endpoints
15 · 18
gateways
openrouter
price rank
cheapest on 0 of 15 · median cheaper than 42%
off card, median
card over 14
cache hit
66.4%
uptime · p50 · speed
96.1% · 1145ms · 57t/s
data policy
no training · retains on some
#57 Reka
reka · reka.ai
tokens / week
9.18B · 0.1%
models · endpoints
10 · 16
gateways
openrouter
price rank
cheapest on 4 of 16 · median cheaper than 69%
off card, median
21% off over 10
cache hit
68.0%
uptime · p50 · speed
98.1% · 967ms · 58t/s
data policy
no training · no retention
#58 AtlasCloud US
atlas-cloud · atlascloud.ai
tokens / week
9.02B · 0.1%
models · endpoints
41 · 41
gateways
openrouter
price rank
cheapest on 4 of 35 · median cheaper than 57%
off card, median
card over 23
cache hit
68.1%
uptime · p50 · speed
97.9% · 1870ms · 38t/s
data policy
no training · retains on some
#59 DekaLLM ID
dekallm · cloudeka.id
tokens / week
8.96B · 0.1%
models · endpoints
10 · 10
gateways
openrouter
price rank
cheapest on 1 of 10 · median cheaper than 80%
off card, median
40% off over 2
cache hit
69.2%
uptime · p50 · speed
97.2% · 593ms · 81t/s
data policy
no training · no retention
#60 Liquid
liquid · liquid.ai
tokens / week
8.76B · 0.1%
models · endpoints
7 · 9
gateways
openrouter vercel
price rank
–
off card, median
–
cache hit
1.1%
uptime · p50 · speed
97.4% · 209ms · 110t/s
data policy
trains on some · retains on some
#61 VoyageAI by MongoDB
voyageai · voyageai.com
tokens / week
8.06B · 0.1%
models · endpoints
16 · 16
gateways
openrouter
price rank
–
off card, median
–
cache hit
0.0%
uptime · p50 · speed
100.0% · – · –
data policy
no training · retains on some
#62 Near AI
near-ai · near.ai
tokens / week
6.96B · 0.0%
models · endpoints
4 · 4
gateways
openrouter
price rank
cheapest on 1 of 4 · median cheaper than 85%
off card, median
card over 2
cache hit
67.5%
uptime · p50 · speed
100.0% · 672ms · 58t/s
data policy
no training · no retention
#63 Groq US
groq · groq.com
tokens / week
6.02B · 0.0%
models · endpoints
10 · 18
gateways
hf openrouter requesty vercel
price rank
cheapest on 0 of 16 · median cheaper than 18%
off card, median
+100% over 1
cache hit
43.0%
uptime · p50 · speed
97.8% · 266ms · 421t/s
data policy
no training · no retention
#64 KT KR
kt · kt.com
tokens / week
5.31B · 0.0%
models · endpoints
7 · 7
gateways
openrouter
price rank
cheapest on 0 of 5 · median cheaper than 55%
off card, median
card over 3
cache hit
66.0%
uptime · p50 · speed
97.9% · 620ms · 116t/s
data policy
no training · retains on some
#65 AkashML
akashml · akashml.com
tokens / week
4.89B · 0.0%
models · endpoints
12 · 15
gateways
openrouter
price rank
cheapest on 3 of 15 · median cheaper than 75%
off card, median
45% off over 5
cache hit
55.4%
uptime · p50 · speed
99.3% · 645ms · 63t/s
data policy
no training · no retention
#66 Io Net US
io-net · io.net
tokens / week
4.81B · 0.0%
models · endpoints
15 · 16
gateways
openrouter
price rank
cheapest on 2 of 16 · median cheaper than 55%
off card, median
27% off over 9
cache hit
63.8%
uptime · p50 · speed
96.8% · 1075ms · 29t/s
data policy
no training · no retention
#67 River US
river · river.ai
tokens / week
2.99B · 0.0%
models · endpoints
4 · 4
gateways
openrouter
price rank
cheapest on 0 of 4 · median cheaper than 67%
off card, median
20% off over 4
cache hit
87.8%
uptime · p50 · speed
95.9% · 852ms · 94t/s
data policy
no training · no retention
#68 Chutes US
chutes · chutes.ai
tokens / week
2.95B · 0.0%
models · endpoints
9 · 9
gateways
openrouter
price rank
cheapest on 3 of 9 · median cheaper than 90%
off card, median
36% off over 4
cache hit
72.3%
uptime · p50 · speed
97.0% · 2213ms · 34t/s
data policy
no training · retains on some
#69 Inception
inception · inceptionlabs.ai
tokens / week
2.81B · 0.0%
models · endpoints
4 · 6
gateways
openrouter vercel
price rank
–
off card, median
–
cache hit
16.4%
uptime · p50 · speed
100.0% · 602ms · 214t/s
data policy
no training · no retention
#70 Cerebras US
cerebras · cerebras.ai
tokens / week
2.76B · 0.0%
models · endpoints
8 · 11
gateways
hf openrouter vercel
price rank
cheapest on 0 of 11 · median cheaper than 0%
off card, median
+275% over 1
cache hit
64.9%
uptime · p50 · speed
100.0% · 269ms · 895t/s
data policy
no training · retains on some
#71 Poolside
poolside · poolside.ai
tokens / week
2.53B · 0.0%
models · endpoints
5 · 6
gateways
openrouter requesty vercel
price rank
–
off card, median
–
cache hit
95.9%
uptime · p50 · speed
99.3% · 573ms · 102t/s
data policy
trains on some · retains on some
#72 PrimaLabs US
primalabs · primalabs.ai
tokens / week
2.25B · 0.0%
models · endpoints
3 · 3
gateways
openrouter
price rank
cheapest on 0 of 3 · median cheaper than 33%
off card, median
10% off over 3
cache hit
83.2%
uptime · p50 · speed
98.6% · 1261ms · 103t/s
data policy
no training · no retention
#73 Ambient
ambient
tokens / week
2.22B · 0.0%
models · endpoints
5 · 6
gateways
openrouter
price rank
cheapest on 0 of 4 · median cheaper than 85%
off card, median
36% off over 4
cache hit
49.5%
uptime · p50 · speed
79.0% · 931ms · 56t/s
data policy
no training · retains on some
#74 Darkbloom
darkbloom · darkbloom.dev
tokens / week
2.13B · 0.0%
models · endpoints
12 · 13
gateways
openrouter
price rank
cheapest on 6 of 11 · median cheapest
off card, median
29% off over 2
cache hit
12.7%
uptime · p50 · speed
97.2% · 2779ms · 56t/s
data policy
no training · retains on some
#75 Tenstorrent US
tenstorrent · tenstorrent.com
tokens / week
1.34B · 0.0%
models · endpoints
5 · 5
gateways
openrouter
price rank
cheapest on 0 of 5 · median cheaper than 69%
off card, median
50% off over 3
cache hit
61.4%
uptime · p50 · speed
97.7% · 924ms · 82t/s
data policy
no training · no retention
#76 Seed SG
seed · byteplus.com
tokens / week
1.31B · 0.0%
models · endpoints
16 · 16
gateways
openrouter
price rank
–
off card, median
–
cache hit
1.8%
uptime · p50 · speed
99.0% · 20631ms · 0t/s
data policy
no training · retains on some
#77 ScaleDown
scaledown · scaledown.ai
tokens / week
1.06B · 0.0%
models · endpoints
6 · 6
gateways
openrouter vercel
price rank
–
off card, median
–
cache hit
0.0%
uptime · p50 · speed
97.6% · 444ms · 140t/s
data policy
trains on some · retains on some
#78 Respan
respan · respan.ai
tokens / week
968.1M · 0.0%
models · endpoints
3 · 3
gateways
openrouter
price rank
–
off card, median
–
cache hit
0.0%
uptime · p50 · speed
100.0% · 356ms · 0t/s
data policy
no training · retains on some
#79 Nex AGI CN
nex-agi · nex-agi.cn
tokens / week
928.3M · 0.0%
models · endpoints
5 · 5
gateways
openrouter
price rank
cheapest on 1 of 1 · median cheapest
off card, median
–
cache hit
79.7%
uptime · p50 · speed
99.8% · 1162ms · 60t/s
data policy
no training · retains on some
#80 Featherless US
featherless · featherless.ai
tokens / week
662.2M · 0.0%
models · endpoints
1 · 1
gateways
openrouter
price rank
cheapest on 1 of 1 · median cheapest
off card, median
card over 1
cache hit
99.6%
uptime · p50 · speed
35.9% · 6413ms · 21t/s
data policy
no training · no retention
#81 Mara US
mara · mara.com
tokens / week
562.3M · 0.0%
models · endpoints
5 · 5
gateways
openrouter
price rank
cheapest on 0 of 5 · median cheaper than 29%
off card, median
+100% over 2
cache hit
36.1%
uptime · p50 · speed
86.2% · 1058ms · 94t/s
data policy
no training · no retention
#82 Mancer 2
mancer · mancer.tech
tokens / week
372.3M · 0.0%
models · endpoints
10 · 11
gateways
openrouter
price rank
cheapest on 1 of 9 · median cheaper than 43%
off card, median
32% off over 5
cache hit
52.2%
uptime · p50 · speed
99.5% · 876ms · 45t/s
data policy
no training · no retention
#83 SambaNova US
sambanova · sambanova.ai
tokens / week
344.8M · 0.0%
models · endpoints
9 · 10
gateways
openrouter vercel
price rank
cheapest on 0 of 10 · median cheaper than 17%
off card, median
card over 5
cache hit
11.2%
uptime · p50 · speed
67.5% · 1774ms · 75t/s
data policy
no training · no retention
#84 AionLabs IL
aion-labs · aionlabs.ai
tokens / week
316.8M · 0.0%
models · endpoints
6 · 6
gateways
openrouter
price rank
–
off card, median
–
cache hit
70.0%
uptime · p50 · speed
99.8% · 1363ms · 52t/s
data policy
no training · retains on some
#85 Ant Digital SG
antdigital · aliyuncs.com
tokens / week
130.4M · 0.0%
models · endpoints
1 · 1
gateways
openrouter
price rank
–
off card, median
–
cache hit
63.7%
uptime · p50 · speed
100.0% · 595ms · 135t/s
data policy
no training · no retention
#86 Unbiased
unbiased · unbiased.ai
tokens / week
129.9M · 0.0%
models · endpoints
2 · 2
gateways
openrouter
price rank
–
off card, median
–
cache hit
85.5%
uptime · p50 · speed
100.0% · 2262ms · 60t/s
data policy
no training · retains on some
#87 Arcee AI US
arcee-ai
tokens / week
119.2M · 0.0%
models · endpoints
2 · 3
gateways
openrouter vercel
price rank
cheapest on 0 of 1 · median cheaper than 0%
off card, median
–
cache hit
94.7%
uptime · p50 · speed
99.3% · 211ms · 84t/s
data policy
no training · retains on some
#88 Black Forest Labs
black-forest-labs · bfl.ai
tokens / week
58.9M · 0.0%
models · endpoints
9 · 9
gateways
openrouter
price rank
–
off card, median
–
cache hit
0.0%
uptime · p50 · speed
100.0% · 27515ms · 0t/s
data policy
no training · retains on some
#89 SambaNova US
sambanova-turbo
tokens / week
54.4M · 0.0%
models · endpoints
1 · 1
gateways
openrouter
price rank
cheapest on 0 of 1 · median cheaper than 33%
off card, median
–
cache hit
0.0%
uptime · p50 · speed
99.8% · 1010ms · 95t/s
data policy
no training · no retention
#90 ParetoAlpha US
pareto-alpha · paretoalphainference.com
tokens / week
47.5M · 0.0%
models · endpoints
1 · 1
gateways
openrouter
price rank
cheapest on 0 of 1 · median cheaper than 72%
off card, median
67% off over 1
cache hit
7.7%
uptime · p50 · speed
96.9% · – · –
data policy
no training · no retention
#91 Sakana AI
sakana · sakana.ai
tokens / week
16.5M · 0.0%
models · endpoints
5 · 9
gateways
openrouter requesty vercel
price rank
–
off card, median
–
cache hit
29.9%
uptime · p50 · speed
100.0% · 4187ms · 98t/s
data policy
trains on some · retains on some
#92 Recraft
recraft
tokens / week
13.6M · 0.0%
models · endpoints
17 · 25
gateways
openrouter vercel
price rank
–
off card, median
–
cache hit
0.0%
uptime · p50 · speed
100.0% · 10283ms · 0t/s
data policy
no training · retains on some
#93 Sourceful
sourceful · sourceful.com
tokens / week
2.7M · 0.0%
models · endpoints
5 · 5
gateways
openrouter
price rank
–
off card, median
–
cache hit
0.0%
uptime · p50 · speed
100.0% · 115612ms · 0t/s
data policy
no training · retains on some
#94 Krea US
krea · krea.ai
tokens / week
2.6M · 0.0%
models · endpoints
4 · 4
gateways
openrouter
price rank
–
off card, median
–
cache hit
0.0%
uptime · p50 · speed
100.0% · 21267ms · 0t/s
data policy
no training · no retention
#95 Fish Audio
fish-audio · fish.audio
tokens / week
1.9M · 0.0%
models · endpoints
10 · 15
gateways
openrouter vercel
price rank
–
off card, median
–
cache hit
0.0%
uptime · p50 · speed
100.0% · – · –
data policy
trains on some · retains on some
#96 ElevenLabs
elevenlabs · elevenlabs.io
tokens / week
1.7M · 0.0%
models · endpoints
17 · 25
gateways
openrouter vercel
price rank
–
off card, median
–
cache hit
0.0%
uptime · p50 · speed
100.0% · – · –
data policy
trains on some · retains on some
#97 Perceptron US
perceptron · perceptron.inc
tokens / week
1.3M · 0.0%
models · endpoints
2 · 2
gateways
openrouter
price rank
–
off card, median
–
cache hit
0.0%
uptime · p50 · speed
100.0% · 758ms · 122t/s
data policy
no training · no retention
#98 HeyGen US
heygen · heygen.com
tokens / week
301903 · 0.0%
models · endpoints
4 · 5
gateways
openrouter vercel
price rank
–
off card, median
–
cache hit
0.0%
uptime · p50 · speed
100.0% · 58875ms · 0t/s
data policy
trains on some · retains on some
#99 Deepgram
deepgram · deepgram.com
tokens / week
62244 · 0.0%
models · endpoints
3 · 3
gateways
openrouter
price rank
–
off card, median
–
cache hit
0.0%
uptime · p50 · speed
100.0% · – · –
data policy
no training · no retention
#100 AssemblyAI
assemblyai · assemblyai.com
tokens / week
2752 · 0.0%
models · endpoints
3 · 4
gateways
openrouter
price rank
cheapest on 2 of 2 · median cheapest
off card, median
–
cache hit
0.0%
uptime · p50 · speed
100.0% · – · –
data policy
no training · retains on some
#101 vertex
vertex
tokens / week
0 · 0.0%
models · endpoints
77 · 92
gateways
requesty vercel
price rank
cheapest on 65 of 75 · median cheapest
off card, median
–
cache hit
–
uptime · p50 · speed
99.9% · 1332ms · 126t/s
data policy
no training · retains on some
#102 featherless-ai
featherless-ai
tokens / week
0 · 0.0%
models · endpoints
79 · 79
gateways
hf
price rank
–
off card, median
–
cache hit
–
uptime · p50 · speed
– · – · –
data policy
not published
#103 bedrock
bedrock
tokens / week
0 · 0.0%
models · endpoints
72 · 75
gateways
requesty vercel
price rank
cheapest on 47 of 65 · median cheapest
off card, median
–
cache hit
–
uptime · p50 · speed
99.8% · 968ms · 105t/s
data policy
no training · retains on some
#104 google
google
tokens / week
0 · 0.0%
models · endpoints
40 · 51
gateways
requesty vercel
price rank
cheapest on 41 of 41 · median cheapest
off card, median
–
cache hit
–
uptime · p50 · speed
95.4% · 1263ms · 210t/s
data policy
no training · retains on some
#105 openai-responses
openai-responses
tokens / week
0 · 0.0%
models · endpoints
37 · 37
gateways
requesty
price rank
cheapest on 33 of 35 · median cheapest
off card, median
–
cache hit
–
uptime · p50 · speed
– · – · –
data policy
no training · retains on some
#106 togetherai
togetherai
tokens / week
0 · 0.0%
models · endpoints
23 · 32
gateways
vercel
price rank
cheapest on 5 of 32 · median cheaper than 50%
off card, median
–
cache hit
–
uptime · p50 · speed
96.1% · 301ms · 123t/s
data policy
no training · no retention
#107 zai
zai
tokens / week
0 · 0.0%
models · endpoints
19 · 27
gateways
requesty vercel
price rank
cheapest on 5 of 16 · median cheaper than 64%
off card, median
–
cache hit
–
uptime · p50 · speed
99.8% · 3459ms · 53t/s
data policy
no training · no retention
#108 azure
azure@eastus2
tokens / week
0 · 0.0%
models · endpoints
23 · 23
gateways
requesty
price rank
cheapest on 22 of 23 · median cheapest
off card, median
–
cache hit
–
uptime · p50 · speed
– · 2420ms · 106t/s
data policy
no training · no retention
#109 zai-org
zai-org
tokens / week
0 · 0.0%
models · endpoints
21 · 21
gateways
hf
price rank
–
off card, median
–
cache hit
–
uptime · p50 · speed
– · 1221ms · 44t/s
data policy
not published
#110 runware
runware
tokens / week
0 · 0.0%
models · endpoints
12 · 20
gateways
requesty vercel
price rank
cheapest on 5 of 18 · median cheaper than 84%
off card, median
–
cache hit
–
uptime · p50 · speed
99.8% · 1294ms · 97t/s
data policy
trains on some · retains on some
#111 fireworks-ai
fireworks-ai
tokens / week
0 · 0.0%
models · endpoints
20 · 20
gateways
hf
price rank
cheapest on 1 of 10 · median cheaper than 50%
off card, median
–
cache hit
–
uptime · p50 · speed
– · 668ms · 67t/s
data policy
not published
#112 nscale
nscale
tokens / week
0 · 0.0%
models · endpoints
20 · 20
gateways
hf
price rank
cheapest on 5 of 10 · median cheaper than 75%
off card, median
–
cache hit
–
uptime · p50 · speed
– · 685ms · 121t/s
data policy
not published
#113 vertexAnthropic
vertexAnthropic
tokens / week
0 · 0.0%
models · endpoints
18 · 18
gateways
vercel
price rank
cheapest on 17 of 17 · median cheapest
off card, median
–
cache hit
–
uptime · p50 · speed
83.3% · 735ms · 90t/s
data policy
no training · retains on some
#114 azure
azure@swedencentral
tokens / week
0 · 0.0%
models · endpoints
18 · 18
gateways
requesty
price rank
cheapest on 1 of 18 · median cheaper than 57%
off card, median
–
cache hit
–
uptime · p50 · speed
– · 2420ms · 121t/s
data policy
no training · no retention
#115 azure
azure@francecentral
tokens / week
0 · 0.0%
models · endpoints
17 · 17
gateways
requesty
price rank
cheapest on 1 of 17 · median cheaper than 60%
off card, median
–
cache hit
–
uptime · p50 · speed
– · 2420ms · 121t/s
data policy
no training · no retention
#116 vertex
vertex@eu
tokens / week
0 · 0.0%
models · endpoints
17 · 17
gateways
requesty
price rank
cheapest on 1 of 17 · median cheaper than 50%
off card, median
–
cache hit
–
uptime · p50 · speed
– · 2175ms · 118t/s
data policy
no training · retains on some
#117 azure
azure@germanywestcentral
tokens / week
0 · 0.0%
models · endpoints
16 · 16
gateways
requesty
price rank
cheapest on 0 of 16 · median cheaper than 57%
off card, median
–
cache hit
–
uptime · p50 · speed
– · 2420ms · 121t/s
data policy
no training · no retention
#118 tensorx
tensorx
tokens / week
0 · 0.0%
models · endpoints
16 · 16
gateways
requesty
price rank
cheapest on 1 of 15 · median cheaper than 14%
off card, median
–
cache hit
–
uptime · p50 · speed
– · 1276ms · 151t/s
data policy
no training · no retention
#119 azure
azure@westeurope
tokens / week
0 · 0.0%
models · endpoints
16 · 16
gateways
requesty
price rank
cheapest on 0 of 16 · median cheaper than 57%
off card, median
–
cache hit
–
uptime · p50 · speed
– · 2420ms · 121t/s
data policy
no training · no retention
#120 bedrock
bedrock@eu-north-1
tokens / week
0 · 0.0%
models · endpoints
14 · 14
gateways
requesty
price rank
cheapest on 1 of 14 · median cheaper than 73%
off card, median
–
cache hit
–
uptime · p50 · speed
– · 1913ms · 106t/s
data policy
no training · no retention
#121 bedrock
bedrock@us-east-1
tokens / week
0 · 0.0%
models · endpoints
14 · 14
gateways
requesty
price rank
cheapest on 2 of 14 · median cheaper than 73%
off card, median
–
cache hit
–
uptime · p50 · speed
– · 1724ms · 78t/s
data policy
no training · retains on some
#122 bytedance
bytedance
tokens / week
0 · 0.0%
models · endpoints
14 · 14
gateways
vercel
price rank
–
off card, median
–
cache hit
–
uptime · p50 · speed
99.9% · 1424ms · 63t/s
data policy
no training · retains on some
#123 voyage
voyage
tokens / week
0 · 0.0%
models · endpoints
14 · 14
gateways
vercel
price rank
–
off card, median
–
cache hit
–
uptime · p50 · speed
100.0% · – · –
data policy
not published
#124 bedrock
bedrock@eu-central-1
tokens / week
0 · 0.0%
models · endpoints
13 · 13
gateways
requesty
price rank
cheapest on 1 of 13 · median cheaper than 73%
off card, median
–
cache hit
–
uptime · p50 · speed
– · 1913ms · 106t/s
data policy
no training · no retention
#125 bedrock
bedrock@us-east-2
tokens / week
0 · 0.0%
models · endpoints
13 · 13
gateways
requesty
price rank
cheapest on 2 of 13 · median cheaper than 75%
off card, median
–
cache hit
–
uptime · p50 · speed
– · 1724ms · 78t/s
data policy
no training · no retention
#126 bfl
bfl
tokens / week
0 · 0.0%
models · endpoints
13 · 13
gateways
vercel
price rank
–
off card, median
–
cache hit
–
uptime · p50 · speed
98.8% · – · –
data policy
not published
#127 bedrock
bedrock@eu-west-1
tokens / week
0 · 0.0%
models · endpoints
13 · 13
gateways
requesty
price rank
cheapest on 1 of 13 · median cheaper than 73%
off card, median
–
cache hit
–
uptime · p50 · speed
– · 1913ms · 106t/s
data policy
no training · no retention
#128 scaleway
scaleway
tokens / week
0 · 0.0%
models · endpoints
11 · 12
gateways
hf requesty
price rank
cheapest on 0 of 10 · median cheaper than 0%
off card, median
–
cache hit
–
uptime · p50 · speed
– · 784ms · 65t/s
data policy
no training · no retention
#129 runinfra
runinfra
tokens / week
0 · 0.0%
models · endpoints
12 · 12
gateways
vercel
price rank
cheapest on 4 of 11 · median cheaper than 83%
off card, median
–
cache hit
–
uptime · p50 · speed
69.7% · 657ms · 122t/s
data policy
trains on some · retains on some
#130 doubleword
doubleword
tokens / week
0 · 0.0%
models · endpoints
12 · 12
gateways
requesty
price rank
cheapest on 0 of 9 · median cheaper than 64%
off card, median
–
cache hit
–
uptime · p50 · speed
– · 1529ms · 1440t/s
data policy
no training · retains on some
#131 vertex
vertex@us-east5
tokens / week
0 · 0.0%
models · endpoints
12 · 12
gateways
requesty
price rank
cheapest on 6 of 12 · median cheaper than 75%
off card, median
–
cache hit
–
uptime · p50 · speed
– · 929ms · 96t/s
data policy
no training · no retention
#132 bedrock
bedrock@eu-west-3
tokens / week
0 · 0.0%
models · endpoints
11 · 11
gateways
requesty
price rank
cheapest on 1 of 11 · median cheaper than 73%
off card, median
–
cache hit
–
uptime · p50 · speed
– · 1755ms · 121t/s
data policy
no training · no retention
#133 bedrock
bedrock@us-west-2
tokens / week
0 · 0.0%
models · endpoints
11 · 11
gateways
requesty
price rank
cheapest on 2 of 11 · median cheaper than 75%
off card, median
–
cache hit
–
uptime · p50 · speed
– · 1314ms · 78t/s
data policy
no training · no retention
#134 boundless
boundless
tokens / week
0 · 0.0%
models · endpoints
10 · 10
gateways
vercel
price rank
cheapest on 1 of 9 · median cheaper than 70%
off card, median
–
cache hit
–
uptime · p50 · speed
100.0% · 581ms · 102t/s
data policy
trains on some · retains on some
#135 vertex
vertex@europe-west1
tokens / week
0 · 0.0%
models · endpoints
10 · 10
gateways
requesty
price rank
cheapest on 5 of 10 · median cheaper than 75%
off card, median
–
cache hit
–
uptime · p50 · speed
– · 1019ms · 113t/s
data policy
no training · no retention
#136 publicai
publicai
tokens / week
0 · 0.0%
models · endpoints
10 · 10
gateways
hf
price rank
–
off card, median
–
cache hit
–
uptime · p50 · speed
– · 1163ms · 56t/s
data policy
not published
#137 vertex
vertex@us
tokens / week
0 · 0.0%
models · endpoints
9 · 9
gateways
requesty
price rank
cheapest on 0 of 9 · median cheaper than 63%
off card, median
–
cache hit
–
uptime · p50 · speed
– · 1776ms · 136t/s
data policy
no training · no retention
#138 nvidia
nvidia · nvidia.com
tokens / week
0 · 0.0%
models · endpoints
9 · 9
gateways
requesty vercel
price rank
–
off card, median
–
cache hit
–
uptime · p50 · speed
99.9% · 1298ms · 70t/s
data policy
trains on some · retains on some
#139 sference
sference
tokens / week
0 · 0.0%
models · endpoints
9 · 9
gateways
requesty
price rank
cheapest on 2 of 8 · median cheaper than 20%
off card, median
–
cache hit
–
uptime · p50 · speed
– · 515ms · 180t/s
data policy
no training · no retention
#140 ovhcloud
ovhcloud
tokens / week
0 · 0.0%
models · endpoints
8 · 8
gateways
hf
price rank
cheapest on 0 of 7 · median cheaper than 50%
off card, median
–
cache hit
–
uptime · p50 · speed
– · 467ms · 43t/s
data policy
not published
#141 klingai
klingai
tokens / week
0 · 0.0%
models · endpoints
8 · 8
gateways
vercel
price rank
–
off card, median
–
cache hit
–
uptime · p50 · speed
100.0% · – · –
data policy
no training · retains on some
#142 claudeaws
claudeaws
tokens / week
0 · 0.0%
models · endpoints
8 · 8
gateways
vercel
price rank
cheapest on 8 of 8 · median cheapest
off card, median
–
cache hit
–
uptime · p50 · speed
100.0% · 694ms · 110t/s
data policy
no training · no retention
#143 azure
azure@polandcentral
tokens / week
0 · 0.0%
models · endpoints
7 · 7
gateways
requesty
price rank
cheapest on 0 of 7 · median cheaper than 67%
off card, median
–
cache hit
–
uptime · p50 · speed
– · 2448ms · 114t/s
data policy
no training · no retention
#144 minimaxi
minimaxi
tokens / week
0 · 0.0%
models · endpoints
7 · 7
gateways
requesty
price rank
cheapest on 3 of 5 · median cheapest
off card, median
–
cache hit
–
uptime · p50 · speed
– · 1473ms · 162t/s
data policy
no training · no retention
#145 particle
particle
tokens / week
0 · 0.0%
models · endpoints
5 · 7
gateways
requesty vercel
price rank
cheapest on 0 of 5 · median cheaper than 70%
off card, median
–
cache hit
–
uptime · p50 · speed
99.7% · 1170ms · 70t/s
data policy
no training · retains on some
#146 melious
melious
tokens / week
0 · 0.0%
models · endpoints
7 · 7
gateways
requesty
price rank
cheapest on 4 of 7 · median cheapest
off card, median
–
cache hit
–
uptime · p50 · speed
– · 649ms · 72t/s
data policy
no training · no retention
#147 moonshot
moonshot
tokens / week
0 · 0.0%
models · endpoints
6 · 6
gateways
requesty
price rank
cheapest on 3 of 5 · median cheapest
off card, median
–
cache hit
–
uptime · p50 · speed
– · – · –
data policy
no training · no retention
#148 lyceum
lyceum
tokens / week
0 · 0.0%
models · endpoints
6 · 6
gateways
requesty
price rank
cheapest on 1 of 6 · median cheaper than 20%
off card, median
–
cache hit
–
uptime · p50 · speed
– · 614ms · 163t/s
data policy
no training · no retention
#149 sail
sail
tokens / week
0 · 0.0%
models · endpoints
5 · 5
gateways
requesty
price rank
cheapest on 1 of 3 · median cheaper than 90%
off card, median
–
cache hit
–
uptime · p50 · speed
– · 4375ms · 81t/s
data policy
no training · no retention
#150 azure
azure@uksouth
tokens / week
0 · 0.0%
models · endpoints
5 · 5
gateways
requesty
price rank
cheapest on 0 of 5 · median cheaper than 44%
off card, median
–
cache hit
–
uptime · p50 · speed
– · 1796ms · 84t/s
data policy
no training · no retention
#151 bedrock
bedrock@ap-northeast-1
tokens / week
0 · 0.0%
models · endpoints
5 · 5
gateways
requesty
price rank
cheapest on 0 of 5 · median cheaper than 73%
off card, median
–
cache hit
–
uptime · p50 · speed
– · 2231ms · 109t/s
data policy
no training · no retention
#152 blackbox
blackbox
tokens / week
0 · 0.0%
models · endpoints
4 · 4
gateways
vercel
price rank
cheapest on 0 of 4 · median cheaper than 75%
off card, median
–
cache hit
–
uptime · p50 · speed
51.1% · 961ms · 94t/s
data policy
trains on some · retains on some
#153 pareto
pareto
tokens / week
0 · 0.0%
models · endpoints
4 · 4
gateways
vercel
price rank
cheapest on 1 of 4 · median cheaper than 87%
off card, median
–
cache hit
–
uptime · p50 · speed
100.0% · 1341ms · 109t/s
data policy
no training · retains on some
#154 vertex
vertex@europe-west8
tokens / week
0 · 0.0%
models · endpoints
4 · 4
gateways
requesty
price rank
cheapest on 4 of 4 · median cheapest
off card, median
–
cache hit
–
uptime · p50 · speed
– · – · –
data policy
no training · no retention
#155 vertex
vertex@europe-west4
tokens / week
0 · 0.0%
models · endpoints
4 · 4
gateways
requesty
price rank
cheapest on 4 of 4 · median cheapest
off card, median
–
cache hit
–
uptime · p50 · speed
– · – · –
data policy
no training · no retention
#156 vertex
vertex@europe-north1
tokens / week
0 · 0.0%
models · endpoints
4 · 4
gateways
requesty
price rank
cheapest on 4 of 4 · median cheapest
off card, median
–
cache hit
–
uptime · p50 · speed
– · – · –
data policy
no training · no retention
#157 thinkingmachines
thinkingmachines · thinkingmachines.ai
tokens / week
0 · 0.0%
models · endpoints
3 · 4
gateways
requesty vercel
price rank
cheapest on 2 of 2 · median cheapest
off card, median
–
cache hit
–
uptime · p50 · speed
100.0% · 1813ms · 165t/s
data policy
trains on some · retains on some
#158 vertex
vertex@us-south1
tokens / week
0 · 0.0%
models · endpoints
4 · 4
gateways
requesty
price rank
cheapest on 4 of 4 · median cheapest
off card, median
–
cache hit
–
uptime · p50 · speed
– · – · –
data policy
no training · no retention
#159 vertex
vertex@us-west1
tokens / week
0 · 0.0%
models · endpoints
4 · 4
gateways
requesty
price rank
cheapest on 4 of 4 · median cheapest
off card, median
–
cache hit
–
uptime · p50 · speed
– · – · –
data policy
no training · no retention
#160 vertex
vertex@europe-central2
tokens / week
0 · 0.0%
models · endpoints
4 · 4
gateways
requesty
price rank
cheapest on 4 of 4 · median cheapest
off card, median
–
cache hit
–
uptime · p50 · speed
– · – · –
data policy
no training · no retention
#161 vertex
vertex@us-central1
tokens / week
0 · 0.0%
models · endpoints
4 · 4
gateways
requesty
price rank
cheapest on 4 of 4 · median cheapest
off card, median
–
cache hit
–
uptime · p50 · speed
– · – · –
data policy
no training · no retention
#162 vertex
vertex@us-east1
tokens / week
0 · 0.0%
models · endpoints
4 · 4
gateways
requesty
price rank
cheapest on 4 of 4 · median cheapest
off card, median
–
cache hit
–
uptime · p50 · speed
– · – · –
data policy
no training · no retention
#163 bedrock
bedrock@ap-southeast-2
tokens / week
0 · 0.0%
models · endpoints
4 · 4
gateways
requesty
price rank
cheapest on 0 of 4 · median cheaper than 73%
off card, median
–
cache hit
–
uptime · p50 · speed
– · 2257ms · 100t/s
data policy
no training · no retention
#164 azure
azure@westus3
tokens / week
0 · 0.0%
models · endpoints
4 · 4
gateways
requesty
price rank
cheapest on 4 of 4 · median cheapest
off card, median
–
cache hit
–
uptime · p50 · speed
– · – · –
data policy
no training · no retention
#165 nace-ai
nace-ai
tokens / week
0 · 0.0%
models · endpoints
3 · 3
gateways
vercel
price rank
–
off card, median
–
cache hit
–
uptime · p50 · speed
99.4% · – · –
data policy
trains on some · retains on some
#166 typesafe-ai
typesafe-ai
tokens / week
0 · 0.0%
models · endpoints
3 · 3
gateways
vercel
price rank
cheapest on 1 of 1 · median cheapest
off card, median
–
cache hit
–
uptime · p50 · speed
100.0% · – · –
data policy
no training · no retention
#167 telnyx
telnyx
tokens / week
0 · 0.0%
models · endpoints
3 · 3
gateways
vercel
price rank
cheapest on 0 of 3 · median cheaper than 50%
off card, median
–
cache hit
–
uptime · p50 · speed
99.6% · 511ms · 253t/s
data policy
trains on some · retains on some
#168 quiverai
quiverai
tokens / week
0 · 0.0%
models · endpoints
3 · 3
gateways
vercel
price rank
–
off card, median
–
cache hit
–
uptime · p50 · speed
96.3% · 1587ms · –
data policy
trains on some · retains on some
#169 topaz
topaz
tokens / week
0 · 0.0%
models · endpoints
3 · 3
gateways
vercel
price rank
–
off card, median
–
cache hit
–
uptime · p50 · speed
100.0% · – · –
data policy
trains on some · retains on some
#170 coding
coding@us-central1
tokens / week
0 · 0.0%
models · endpoints
2 · 2
gateways
requesty
price rank
cheapest on 2 of 2 · median cheapest
off card, median
–
cache hit
–
uptime · p50 · speed
– · – · –
data policy
no training · no retention
#171 bedrock
bedrock@ap-northeast-3
tokens / week
0 · 0.0%
models · endpoints
2 · 2
gateways
requesty
price rank
cheapest on 0 of 2 · median cheaper than 73%
off card, median
–
cache hit
–
uptime · p50 · speed
– · 3585ms · 121t/s
data policy
no training · no retention
#172 bedrock
bedrock@ap-south-1
tokens / week
0 · 0.0%
models · endpoints
2 · 2
gateways
requesty
price rank
cheapest on 0 of 2 · median cheaper than 73%
off card, median
–
cache hit
–
uptime · p50 · speed
– · 2049ms · 87t/s
data policy
no training · no retention
#173 Parallel
parallel · parallel.ai
tokens / week
0 · 0.0%
models · endpoints
2 · 2
gateways
openrouter vercel
price rank
–
off card, median
–
cache hit
–
uptime · p50 · speed
99.7% · 547ms · 0t/s
data policy
no training · retains on some
#174 prodia
prodia
tokens / week
0 · 0.0%
models · endpoints
2 · 2
gateways
vercel
price rank
–
off card, median
–
cache hit
–
uptime · p50 · speed
98.6% · – · –
data policy
no training · retains on some
#175 reflection
reflection
tokens / week
0 · 0.0%
models · endpoints
2 · 2
gateways
vercel
price rank
–
off card, median
–
cache hit
–
uptime · p50 · speed
99.3% · 284ms · 131t/s
data policy
trains on some · retains on some
#176 bedrock
bedrock@eu-west-2
tokens / week
0 · 0.0%
models · endpoints
2 · 2
gateways
requesty
price rank
cheapest on 0 of 2 · median cheaper than 20%
off card, median
–
cache hit
–
uptime · p50 · speed
– · 1913ms · 101t/s
data policy
no training · no retention
#177 Runway US
runway · runwayml.com
tokens / week
0 · 0.0%
models · endpoints
2 · 2
gateways
openrouter
price rank
–
off card, median
–
cache hit
–
uptime · p50 · speed
– · 132663ms · 0t/s
data policy
no training · retains on some
#178 coding
coding@us-west1
tokens / week
0 · 0.0%
models · endpoints
2 · 2
gateways
requesty
price rank
cheapest on 2 of 2 · median cheapest
off card, median
–
cache hit
–
uptime · p50 · speed
– · – · –
data policy
no training · no retention
#179 coding
coding@us-south1
tokens / week
0 · 0.0%
models · endpoints
2 · 2
gateways
requesty
price rank
cheapest on 2 of 2 · median cheapest
off card, median
–
cache hit
–
uptime · p50 · speed
– · – · –
data policy
no training · no retention
#180 coding
coding@us-east5
tokens / week
0 · 0.0%
models · endpoints
2 · 2
gateways
requesty
price rank
cheapest on 2 of 2 · median cheapest
off card, median
–
cache hit
–
uptime · p50 · speed
– · – · –
data policy
no training · no retention
#181 coding
coding@us-east1
tokens / week
0 · 0.0%
models · endpoints
2 · 2
gateways
requesty
price rank
cheapest on 2 of 2 · median cheapest
off card, median
–
cache hit
–
uptime · p50 · speed
– · – · –
data policy
no training · no retention
#182 subconscious
subconscious
tokens / week
0 · 0.0%
models · endpoints
2 · 2
gateways
vercel
price rank
cheapest on 0 of 2 · median cheaper than 43%
off card, median
–
cache hit
–
uptime · p50 · speed
99.9% · 411ms · 90t/s
data policy
trains on some · retains on some
#183 coding
coding@europe-west8
tokens / week
0 · 0.0%
models · endpoints
2 · 2
gateways
requesty
price rank
cheapest on 2 of 2 · median cheapest
off card, median
–
cache hit
–
uptime · p50 · speed
– · – · –
data policy
no training · no retention
#184 coding
coding@europe-west4
tokens / week
0 · 0.0%
models · endpoints
2 · 2
gateways
requesty
price rank
cheapest on 2 of 2 · median cheapest
off card, median
–
cache hit
–
uptime · p50 · speed
– · – · –
data policy
no training · no retention
#185 coding
coding@europe-west1
tokens / week
0 · 0.0%
models · endpoints
2 · 2
gateways
requesty
price rank
cheapest on 2 of 2 · median cheapest
off card, median
–
cache hit
–
uptime · p50 · speed
– · – · –
data policy
no training · no retention
#186 coding
coding@europe-north1
tokens / week
0 · 0.0%
models · endpoints
2 · 2
gateways
requesty
price rank
cheapest on 2 of 2 · median cheapest
off card, median
–
cache hit
–
uptime · p50 · speed
– · – · –
data policy
no training · no retention
#187 coding
coding@europe-central2
tokens / week
0 · 0.0%
models · endpoints
2 · 2
gateways
requesty
price rank
cheapest on 2 of 2 · median cheapest
off card, median
–
cache hit
–
uptime · p50 · speed
– · – · –
data policy
no training · no retention
#188 coding
coding
tokens / week
0 · 0.0%
models · endpoints
2 · 2
gateways
requesty
price rank
cheapest on 2 of 2 · median cheapest
off card, median
–
cache hit
–
uptime · p50 · speed
– · – · –
data policy
no training · no retention
#189 vertex
vertex@europe-southwest1
tokens / week
0 · 0.0%
models · endpoints
1 · 1
gateways
requesty
price rank
cheapest on 1 of 1 · median cheapest
off card, median
–
cache hit
–
uptime · p50 · speed
– · – · –
data policy
no training · no retention
#190 vertex
vertex@asia-northeast1
tokens / week
0 · 0.0%
models · endpoints
1 · 1
gateways
requesty
price rank
cheapest on 1 of 1 · median cheapest
off card, median
–
cache hit
–
uptime · p50 · speed
– · – · –
data policy
no training · no retention
#191 AI21 IL
ai21 · ai21.com
tokens / week
0 · 0.0%
models · endpoints
1 · 1
gateways
openrouter
price rank
–
off card, median
–
cache hit
–
uptime · p50 · speed
– · – · –
data policy
no training · retains on some
#192 vertex
vertex@us-east4
tokens / week
0 · 0.0%
models · endpoints
1 · 1
gateways
requesty
price rank
cheapest on 1 of 1 · median cheapest
off card, median
–
cache hit
–
uptime · p50 · speed
– · – · –
data policy
no training · no retention
#193 azure
azure@eastus
tokens / week
0 · 0.0%
models · endpoints
1 · 1
gateways
requesty
price rank
cheapest on 1 of 1 · median cheapest
off card, median
–
cache hit
–
uptime · p50 · speed
– · 2584ms · 150t/s
data policy
no training · no retention
#194 azure
azure@japaneast
tokens / week
0 · 0.0%
models · endpoints
1 · 1
gateways
requesty
price rank
cheapest on 0 of 1 · median cheaper than 82%
off card, median
–
cache hit
–
uptime · p50 · speed
– · 2448ms · 82t/s
data policy
no training · no retention
#195 elorian
elorian
tokens / week
0 · 0.0%
models · endpoints
1 · 1
gateways
vercel
price rank
–
off card, median
–
cache hit
–
uptime · p50 · speed
100.0% · – · –
data policy
trains on some · retains on some
#196 vertex
vertex@us-west4
tokens / week
0 · 0.0%
models · endpoints
1 · 1
gateways
requesty
price rank
cheapest on 1 of 1 · median cheapest
off card, median
–
cache hit
–
uptime · p50 · speed
– · – · –
data policy
no training · no retention
#197 bedrock
bedrock@eu-south-1
tokens / week
0 · 0.0%
models · endpoints
1 · 1
gateways
requesty
price rank
cheapest on 0 of 1 · median cheaper than 33%
off card, median
–
cache hit
–
uptime · p50 · speed
– · – · –
data policy
no training · no retention
#198 azure
azure@koreacentral
tokens / week
0 · 0.0%
models · endpoints
1 · 1
gateways
requesty
price rank
cheapest on 0 of 1 · median cheaper than 82%
off card, median
–
cache hit
–
uptime · p50 · speed
– · 2448ms · 82t/s
data policy
no training · no retention
#199 humansand
humansand
tokens / week
0 · 0.0%
models · endpoints
1 · 1
gateways
vercel
price rank
–
off card, median
–
cache hit
–
uptime · p50 · speed
99.9% · 18545ms · –
data policy
trains on some · retains on some
#200 callstack
callstack
tokens / week
0 · 0.0%
models · endpoints
1 · 1
gateways
vercel
price rank
–
off card, median
–
cache hit
–
uptime · p50 · speed
100.0% · 861ms · –
data policy
trains on some · retains on some
#201 azure-responses
azure-responses@eastus2
tokens / week
0 · 0.0%
models · endpoints
1 · 1
gateways
requesty
price rank
cheapest on 1 of 1 · median cheapest
off card, median
–
cache hit
–
uptime · p50 · speed
– · – · –
data policy
no training · no retention
#202 interfaze
interfaze
tokens / week
0 · 0.0%
models · endpoints
1 · 1
gateways
vercel
price rank
–
off card, median
–
cache hit
–
uptime · p50 · speed
100.0% · 1046ms · 132t/s
data policy
no training · retains on some
#203 meituan
meituan
tokens / week
0 · 0.0%
models · endpoints
1 · 1
gateways
vercel
price rank
–
off card, median
–
cache hit
–
uptime · p50 · speed
99.9% · 12052ms · 44t/s
data policy
not published
#204 mixedbread
mixedbread
tokens / week
0 · 0.0%
models · endpoints
1 · 1
gateways
vercel
price rank
–
off card, median
–
cache hit
–
uptime · p50 · speed
100.0% · 456ms · –
data policy
trains on some · retains on some