Skip to content
ModelCap

Inference provider

DeepInfra: AI models and API pricing

Every tracked model DeepInfra serves through OpenRouter, with DeepInfra's own input and output price per 1M tokens, the context window it serves, disclosed quantization and measured endpoint uptime. Models with a position on a ModelCap board link to their model page, where every provider's deployment is listed side by side.

Snapshot as of 19 November 2026

DeepInfra serves 65 tracked models through 131 endpoints as of 19 November 2026; the highest-ranked is Kimi K3.1 Pro (ModelCap Index #11); mean measured uptime is 98.47%.

Models served
65
On the ModelCap Index
43
Cheapest output price
$0.024/1M

Language models on DeepInfra (65)

Ranked models first, in ModelCap Index order. Prices are the lowest DeepInfra lists for the model; the endpoint count shows how many deployments carry it.

Language models served by DeepInfra, with price, context, quantization and uptime
#ModelModelCap IndexInput $/1MOutput $/1MContextQuantizationUptime (30 min)Endpoints
1Kimi K3.1 ProMoonshot AI#1180.0$1.90$9.501Mbf16100.0%1
2Kimi K3Moonshot AI#1779.0$2.85$14.251Mmxfp4100.0%1
3GLM 5.3Z.ai#1978.5$0.45$2.001Mfp4100.0%2
4MiMo-V2.6-ProXiaomi#2078.4$0.387$0.7831Mfp8100.0%1Open weights
5GLM 5.3 FlashZ.ai#2676.4$0.15$0.501Mfp4100.0%1Open weights
6DeepSeek V4 Pro 0813DeepSeek#3473.0$1.04$2.081Mfp8100.0%3Open weights
7Gemma 4 31BGoogle#4270.7$0.135$0.32262Kfp4, fp8100.0%4Open weights
8Hy4 previewTencent#4569.5$0.834$2.501Mfp4100.0%1Open weights
9MiMo-V2.6-FlashXiaomi#4668.8$0.14$0.281Mfp8100.0%1Open weights
10Qwen3.5 397B A17BQwen#5267.2$0.36$2.40262Kfp8100.0%3Open weights
11Muse Glimmer 30BMeta#5666.8$0.30$1.20131Kbf16100.0%1Open weights
12Gemma 4 26B A4BGoogle#6065.9$0.07$0.34262Kfp8100.0%1Open weights
13MiniMax M3MiniMax#6663.7$0.224$0.88524Kfp8100.0%2
14Qwen3.8 27BQwen#6863.1$0.15$1.88262Kfp8100.0%1Open weights
15DeepSeek V3.2DeepSeek#8958.8$0.26$0.38164Kfp4100.0%1Open weights
16Qwen3 235B A22B Instruct 2507Qwen#9158.5$0.072$0.44262Kfp8100.0%3Open weights
17DeepSeek V4.1 Flash 1017DeepSeek#9258.5$0.078$0.2351Mfp878.89%3Open weights
18R1 0528DeepSeek#9558.1$0.50$2.15164Kfp4100.0%1Open weights
19Inkling 2 SmallThinking Machines#9856.8$0.285$0.951Mfp460.00%1Open weights
20Qwen3 VL 235B A22B InstructQwen#10755.6$0.20$0.88262Kfp8100.0%1Open weights
21Qwen3.6 35B A3BQwen#11652.6$0.08$0.76262Kfp895.46%3Open weights
22Qwen3 Next 80B A3B InstructQwen#12451.7$0.072$0.88262Kfp8100.0%3Open weights
23Nemotron 3.5 Super VLNVIDIA#13350.9$0.15$0.50262Kfp4100.0%1
24Qwen3 Coder 480B A35BQwen#14248.9$0.30$1.00262Kfp495.83%1Open weights
25Gemma 3 27BGoogle#16042.7$0.064$0.128131Kfp8100.0%3
26Nemotron 3 SuperNVIDIA#16341.7$0.06$0.32262Kbf1697.92%3
27Qwen3.5-9BQwen#16441.1$0.08$0.12262Kbf16100.0%3Open weights
28gpt-oss-120bOpenAI#17439.8$0.037$0.17131Kbf16100.0%1Open weights
29Qwen3 32BQwen#17539.5$0.064$0.22441Kfp8100.0%3Open weights
30Qwen3 VL 30B A3B InstructQwen#18438.7$0.15$0.60262Kfp8100.0%1Open weights
31Gemma 3 12BGoogle#20136.7$0.04$0.12131Kbf16100.0%3
32Nemotron 3.5 LightningNVIDIA#21934.9$0.06$0.16262Kbf16100.0%1
33Qwen3 30B A3BQwen#22234.6$0.096$0.4041Kfp8100.0%3Open weights
34gpt-oss-20bOpenAI#23832.2$0.024$0.112131Kbf1697.83%2Open weights
35Llama 4 ScoutMeta#23931.8$0.08$0.24328Kfp8100.0%3
36Llama 3.3 70B InstructMeta#24431.4$0.10$0.32131Kfp8100.0%1
37Mistral Small 3.2 24BMistral AI#25828.4$0.06$0.16128Kfp8100.0%3Open weights
38Qwen2.5 72B InstructQwen#25928.3$0.288$0.3233Kfp8100.0%2
39Nemotron 3 UltraNVIDIA#26128.0$0.40$1.76262Kfp4100.0%3
40Gemma 3 4BGoogle#26726.8$0.04$0.08131Kbf16100.0%3
41Granite 4.2 8BIBM Granite#26926.3$0.06$0.25131Kbf16100.0%1Open weights
42Phi 4Microsoft#30819.2$0.056$0.11216Kbf16100.0%2Open weights
43Llama 3.1 8B InstructMeta#32513.8$0.02$0.04131Kfp893.33%1
44DeepSeek V4 Flash Vision ExpDeepSeek—$0.172$0.5171Mfp8100.0%3Open weights
45Ling 3.0 Flash VLInclusionAI—$0.06$0.18131Kfp4100.0%1Open weights
46Llama Guard 4 12BMeta—$0.144$0.144164Kbf16100.0%3
47Mistral NemoMistral AI—$0.023$0.024131Kfp8100.0%2Open weights
48Nemotron 3.5 Content SafetyNVIDIA—$0.20$0.20131Kbf16—1
49Qwen3 14BQwen—$0.096$0.19241Kfp898.34%3Open weights
50DeepSeek V3DeepSeek—$0.256$0.712164Kfp4100.0%2
51DeepSeek V3.1DeepSeek—$0.25$0.95164Kfp4100.0%1Open weights
52DeepSeek V4 Flash 0423DeepSeek—$0.072$0.1441Mfp8100.0%3Open weights
53DeepSeek V4 Flash 0731DeepSeek—$0.048$0.1441Mfp892.60%3Open weights
54DeepSeek V4 Pro 0423DeepSeek—$1.04$2.081Mfp8100.0%3Open weights
55DeepSeek V4.1 FlashDeepSeek—$0.112$0.3361Mfp8100.0%3Open weights
56GLM 4.6Z.ai—$0.40$1.60203Kfp4100.0%2Open weights
57GLM 4.7Z.ai—$0.32$1.40203Kfp4100.0%2Open weights
58GLM 5.2Z.ai—$0.45$1.441Mfp4100.0%3Open weights
59Inkling SmallThinking Machines—$0.45$1.20524Kfp4100.0%1Open weights
60Kimi K2.6Moonshot AI—$0.60$2.80262Kfp4100.0%3
61Ling 3.0 FlashInclusionAI—$0.06$0.18131Kfp4100.0%1Open weights
62Mistral Small 3Mistral AI—$0.04$0.06433Kfp8100.0%2Open weights
63Qwen3.5-27BQwen—$0.26$2.60262Kfp8100.0%1Open weights
64Qwen3.5-35B-A3BQwen—$0.112$0.80262Kfp8100.0%2Open weights
65Qwen3.6 27BQwen—$0.256$2.56262Kfp8100.0%3Open weights

DeepInfra on ModelCap: common questions

How many AI models does DeepInfra serve?

65 tracked models as of 19 November 2026: 65 language models (43 with a public ModelCap Index position), across 131 live endpoints observed through OpenRouter.

What is the best model available on DeepInfra?

Kimi K3.1 Pro from Moonshot AI is the highest-ranked language model DeepInfra serves, at ModelCap Index position #11 and $9.50 per 1M output tokens on DeepInfra. The table lists every model in Index order.

What is the cheapest model on DeepInfra?

Mistral Nemo at $0.024 per 1M output tokens ($0.023 per 1M input), the lowest positive language-token price DeepInfra lists. Free-tier endpoints, where present, show as $0 and are excluded from that comparison.

Which DeepInfra model has the largest context window?

Kimi K3.1 Pro, served with a 1M-token context window on DeepInfra. Providers sometimes serve less than the model's published maximum, so the figure here is the provider's own.

How reliable is DeepInfra?

Mean measured uptime across DeepInfra endpoints is 98.47% over the last 30 minutes, as reported by OpenRouter at the time of the snapshot. Per-model uptime is in the table; it is a live operational signal, not a service-level guarantee.

Where do the DeepInfra prices come from?

From the OpenRouter catalogue entry for each DeepInfra endpoint, in USD per 1M tokens, re-observed on every ModelCap refresh. Where DeepInfra exposes several deployments of one model at different prices, the table shows the lowest and notes the count; the model page lists each deployment.

How current is this page?

This page uses the published ModelCap snapshot dated 19 November 2026. Publication and page updates happen separately. Offers retained from a failed refresh are marked stale in the data and are not counted here.

Explore further