Inference providers

LLM API providers compared

79 inference providers serving at least one tracked model through OpenRouter. Uptime is measured across each provider's endpoints over the last 30 minutes; each row averages that provider's deployments, and the summary gives every provider equal weight, so neither is a per-model guarantee. Entry price is the cheapest positive language-token output price the provider lists. Every provider name opens a page listing each model it serves with that provider's own price, context window, quantization and uptime.

Snapshot as of 23 September 2026

Providers

Mean provider uptime

98.35%

Largest catalogue

74 models

Inference providers by number of tracked models served, with entry price and measured uptime
#	Provider	Models served	On the Index	Highest-ranked model	Entry price ($/1M out)	Uptime
1	DeepInfra	74	48	MiMo-V2.6-Pro	$0.03	95.16%
2	Novita	70	39	GLM 5.3	$0.05	98.29%
3	Alibaba	55	24	Qwen3.8 Max (0902)	$0.13	99.72%
4	OpenAI	48	13	GPT-6 Sol	$0.20	100.0%
5	Azure	47	13	Claude Opus 5.5	$0.40	99.69%
6	Google	44	19	Claude Opus 5.5	$0.25	96.16%
7	SiliconFlow	39	20	GLM 5.3	$0.15	95.09%
8	Amazon Bedrock	36	15	Claude Opus 5.5	$0.14	99.27%
9	Venice	34	19	GLM 5.3	$0.15	95.79%
10	Parasail	33	23	GLM 5.3	$0.03	99.75%
11	AtlasCloud	29	12	GLM 5.3	$0.28	98.85%
12	GMICloud	23	10	GLM 5.3	$0.182	96.81%
13	StreamLake	23	13	GLM 5.3 Flash	$0.158	99.11%
14	Google AI Studio	22	3	Gemini 3.8 Flash	$0.20	99.84%
15	Phala	20	12	GLM 5.3	$0.20	98.60%
16	CoreWeave	19	15	GLM 5.3 Flash	$0.13	99.49%
17	Cloudflare	18	11	GLM 5.3	$0.112	90.26%
18	DigitalOcean	16	9	GLM 5.3	$0.196	99.02%
19	Mistral	15	5	GLM 5.3	$0.10	99.66%
20	BaseTen	14	9	GLM 5.3	$0.26	99.22%
21	Z.AI	14	6	GLM 5.3	$0.50	99.64%
22	Together	13	11	GLM 5.3	$0.25	98.30%
23	Anthropic	12	4	Claude Opus 5.5	$5.00	99.93%
24	Fireworks	11	7	GLM 5.3	$0.50	99.63%
25	Baidu	10	3	GLM 5.3	$0.168	99.23%
26	Claude Platform on AWS	10	2	Claude Opus 5.5	$10.00	99.94%
27	DekaLLM	8	7	DeepSeek V4.1 Flash	$0.03	98.66%
28	Minimax	8	2	MiniMax M3	$1.02	97.63%
29	NextBit	8	3	GLM 5.3 Flash	$0.22	99.39%
30	Wafer	8	6	GLM 5.3	$0.35	99.74%
31	Cohere	7	5	Command A+	$0.15	99.33%
32	Darkbloom	7	6	Qwen3.8 27B	$0.09	99.57%
33	Friendli	7	4	GLM 5.3	$0.40	99.86%
34	Morph	7	4	GLM 5.3	$0.308	97.84%
35	Nebius	7	5	Qwen3 235B A22B Instruct 2507	$0.24	93.01%
36	Sail Research	7	5	GLM 5.3	$0.475	99.66%
37	SambaNova	7	5	Gemma 4 31B	$0.90	97.60%
38	xAI	7	2	Grok 4.7	$2.00	98.62%
39	AionLabs	6	0	—	$1.40	100.0%
40	AkashML	6	6	GLM 5.3	$0.10	99.86%
41	Chutes	6	3	Kimi K3	$0.37	94.61%
42	Crusoe	6	5	GLM 5.3	$0.20	97.63%
43	Groq	6	5	gpt-oss-120b	$0.08	99.69%
44	Inceptron	6	3	GLM 5.3	$0.50	99.71%
45	Reka	6	3	GLM 5.3	$0.10	98.97%
46	Relace	6	3	Kimi K3	$0.36	99.91%
47	Seed	6	1	Seed 2.1 Turbo	$0.30	100.0%
48	Makora	5	4	GLM 5.3	$0.195	98.99%
49	Mancer 2	5	2	Qwen3.8 27B	$0.30	98.25%
50	Mara	5	3	MiniMax M3	$0.75	81.22%
51	Meta	5	1	Muse Spark 1.3	$0.20	100.0%
52	Perplexity	5	0	—	$1.00	100.0%
53	Tencent	5	1	Hy4 preview	$0.177	100.0%
54	Xiaomi	5	3	MiMo-V2.6-Pro	$0.28	100.0%
55	Modal	4	4	GLM 5.3	$1.20	98.83%
56	OpenInference	4	2	GLM 5.3 Flash	$0.50	96.54%
57	Sakana AI	4	0	—	$4.00	100.0%
58	Decart	3	1	GLM 5.3	$2.47	99.76%
59	Inception	3	1	Mercury 2.5	$0.15	100.0%
60	InferenceNet	3	3	GLM 5.3	$0.28	98.80%
61	ModelRun	3	3	Gemma 4 31B	$1.00	99.90%
62	Moonshot AI	3	2	Kimi K3	$4.00	99.72%
63	Upstage	3	1	Solar Pro 4	$0.20	99.92%
64	DeepSeek	2	2	DeepSeek V4.1 Flash	$0.60	100.0%
65	Io Net	2	2	GLM 5.3	$0.475	99.82%
66	Ionstream	2	2	DeepSeek V4 Pro 0813	$2.30	99.86%
67	Nex AGI	2	2	Nex-N2.5-Pro	$0.10	100.0%
68	Poolside	2	2	Laguna S 2.1	$0.12	100.0%
69	Arcee AI	1	1	Trinity Large Thinking	$0.80	100.0%
70	Cerebras	1	1	gpt-oss-120b	$0.75	100.0%
71	Krea	1	1	DeepSeek V4.1 Flash	$0.90	99.84%
72	Liquid	1	0	—	—	100.0%
73	Near AI	1	1	GLM 5.3 Flash	$0.50	99.33%
74	Nvidia	1	0	—	—	72.00%
75	Perceptron	1	0	—	$1.50	100.0%
76	PrimeIntellect	1	1	GLM 5.3	$4.40	100.0%
77	Stealth	1	0	—	—	100.0%
78	StepFun	1	1	Step 3.7 Flash	$1.15	99.48%
79	Unbiased	1	0	—	$7.50	100.0%

LLM API providers: common questions

How many LLM API providers does ModelCap track?

79 inference providers with at least one live endpoint for a tracked model as of 23 September 2026, observed through OpenRouter. Each provider name links to a page listing every model it serves with that provider's own price, context window, quantization and uptime.

Which provider serves the most AI models?

DeepInfra, with 74 tracked models across 78 endpoints. Model count is unique canonical models, so a product served from several deployments counts once.

Which provider has the cheapest LLM API pricing?

The lowest positive language-token entry price is $0.03 per 1M output tokens on DeepInfra. Entry price is the cheapest model a provider lists, not a like-for-like comparison; the cheapest-LLMs ranking and the compare tool line up specific models across providers.

Which provider has the best uptime?

AionLabs shows the highest mean measured uptime, 100.0%, over the last 30 minutes at the time of the snapshot; 79 of 79 providers report a measurement. Uptime is per endpoint from OpenRouter and averaged with equal weight, so it is a live signal rather than an SLA.

How are provider prices and uptime measured?

Both come from OpenRouter's per-endpoint catalogue and telemetry, re-observed on every ModelCap refresh. Prices are USD per 1M tokens as listed by the provider; uptime is the provider's own endpoint availability over the last 30 minutes. Offers retained from a failed refresh are marked stale and excluded from every count on this page.