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. Rows link to the model page, where every provider's deployment is listed side by side.

Snapshot as of 1 October 2026

DeepInfra serves 66 tracked models through 68 endpoints as of 1 October 2026; the highest-ranked is MiMo-V2.6-Pro (ModelCap Index #6); mean measured uptime is 98.52%.

Models served

On the ModelCap Index

Cheapest output price

$0.03/1M

Language models on DeepInfra (66)

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
#	Model	ModelCap Index	Input $/1M	Output $/1M	Context	Quantization	Uptime (30 min)	Endpoints
1	MiMo-V2.6-ProXiaomi	#683.1	$0.43	$0.87	1M	fp8	99.43%	1Open weights
2	Kimi K3Moonshot AI	#1082.2	$2.85	$14.25	1M	bf16	84.17%	1
3	GLM 5.3Z.ai	#1180.9	$0.563	$2.50	1M	fp4	99.72%	1
4	DeepSeek V4.1 FlashDeepSeek	#1579.3	$0.14	$0.42	1M	fp8	100.0%	1Open weights
5	GLM 5.3 FlashZ.ai	#1778.9	$0.075	$0.25	1M	fp4	100.0%	1Open weights
6	DeepSeek V4 Pro 0813DeepSeek	#2476.4	$1.30	$2.60	1M	fp8	99.84%	1Open weights
7	MiMo-V2.6-FlashXiaomi	#2575.7	$0.14	$0.28	1M	fp8	98.64%	1Open weights
8	Hy4 previewTencent	#2673.8	$0.834	$2.50	1M	fp8	100.0%	1Open weights
9	DeepSeek V4 Flash Vision ExpDeepSeek	#2773.8	$0.216	$0.647	1M	fp8	99.47%	1Open weights
10	Gemma 4 31BGoogle	#2872.9	$0.09	$0.34	262K	fp4, fp8	99.13%	2Open weights
11	Muse Glimmer 30BMeta	#3470.7	$0.30	$1.20	131K	bf16	100.0%	1Open weights
12	Qwen3.8 27BQwen	#3968.9	$0.15	$1.88	262K	bf16	98.97%	1Open weights
13	MiniMax M3MiniMax	#4068.5	$0.28	$1.10	524K	fp8	99.82%	1
14	Gemma 4 26B A4BGoogle	#4168.4	$0.07	$0.34	262K	fp8	99.53%	1Open weights
15	Qwen3.5 397B A17BQwen	#4368.0	$0.45	$3.00	262K	fp8	100.0%	1Open weights
16	InklingThinking Machines	#4767.5	$0.95	$4.05	524K	fp8	99.42%	1Open weights
17	Ling 3.0 Flash VLInclusionAI	#5065.9	$0.06	$0.18	131K	fp16	100.0%	1Open weights
18	Ling 3.0 Flash FinInclusionAI	#5964.3	$0.06	$0.18	262K	fp4	—	1Open weights
19	Ling 3.0 FlashInclusionAI	#6362.2	$0.06	$0.18	131K	bf16	99.91%	1Open weights
20	Qwen3 235B A22B Instruct 2507Qwen	#6462.1	$0.09	$0.55	262K	fp8	99.00%	1Open weights
21	DeepSeek V3.2DeepSeek	#6661.7	$0.26	$0.38	164K	fp4	99.92%	1Open weights
22	R1 0528DeepSeek	#6761.5	$0.50	$2.15	164K	fp4	99.93%	1Open weights
23	Qwen3.6 35B A3BQwen	#7260.7	$0.10	$0.95	262K	fp8	92.84%	1Open weights
24	Qwen3 VL 235B A22B InstructQwen	#7859.5	$0.20	$0.88	262K	fp8	100.0%	1Open weights
25	Inkling SmallThinking Machines	#8257.6	$0.45	$1.20	524K	fp8	100.0%	1Open weights
26	Nemotron 3.5 LightningNVIDIA	#8756.3	$0.06	$0.16	262K	bf16	100.0%	1
27	Qwen3.5-9BQwen	#9554.9	$0.10	$0.15	262K	bf16	100.0%	1Open weights
28	Qwen3 Next 80B A3B InstructQwen	#9854.7	$0.09	$1.10	262K	fp8	99.73%	1Open weights
29	Qwen3 Coder 480B A35BQwen	#11551.9	$0.30	$1.00	262K	fp4	96.72%	1Open weights
30	gpt-oss-120bOpenAI	#13244.6	$0.037	$0.17	131K	bf16	100.0%	2Open weights
31	Gemma 3 27BGoogle	#13344.5	$0.08	$0.16	131K	fp8	99.78%	1
32	Nemotron 3 SuperNVIDIA	#13444.0	$0.085	$0.40	262K	bf16	100.0%	1
33	Mistral Small 3.2 24BMistral AI	#14143.1	$0.075	$0.20	128K	fp8	100.0%	1Open weights
34	Qwen3 32BQwen	#14441.2	$0.08	$0.28	41K	fp8	100.0%	1Open weights
35	Gemma 3 12BGoogle	#16338.1	$0.05	$0.15	131K	bf16	100.0%	1
36	gpt-oss-20bOpenAI	#16637.8	$0.03	$0.14	131K	bf16	100.0%	1Open weights
37	Qwen3 30B A3BQwen	#17135.7	$0.12	$0.50	41K	fp8	100.0%	1Open weights
38	Granite 4.2 8BIBM Granite	#17234.8	$0.06	$0.25	131K	bf16	100.0%	1Open weights
39	Llama 4 ScoutMeta	#17732.9	$0.10	$0.30	328K	fp8	99.88%	1
40	Llama 3.3 70B InstructMeta	#17832.4	$0.10	$0.32	131K	fp8	98.44%	1
41	Nemotron 3 Nano 30B A3BNVIDIA	#17931.8	$0.05	$0.20	262K	fp4	100.0%	1
42	Qwen2.5 72B InstructQwen	#18929.0	$0.36	$0.40	33K	fp8	99.35%	1
43	Gemma 3 4BGoogle	#19427.4	$0.05	$0.10	131K	bf16	100.0%	1
44	Phi 4Microsoft	#21919.6	$0.07	$0.14	16K	bf16	100.0%	1Open weights
45	Llama 3.1 8B InstructMeta	#23313.5	$0.02	$0.04	131K	fp8	100.0%	1
46	Llama Guard 4 12BMeta	—	$0.18	$0.18	164K	bf16	100.0%	1
47	Mistral NemoMistral AI	—	$0.019	$0.03	131K	fp8	100.0%	1Open weights
48	Nemotron 3 UltraNVIDIA	—	$0.50	$2.20	262K	fp4	—	1
49	Nemotron 3.5 Content SafetyNVIDIA	—	$0.20	$0.20	131K	bf16	—	1
50	Qwen3 14BQwen	—	$0.12	$0.24	41K	fp8	95.67%	1Open weights
51	Qwen3 VL 30B A3B InstructQwen	—	$0.15	$0.60	262K	fp8	99.81%	1Open weights
52	DeepSeek V3DeepSeek	—	$0.32	$0.89	164K	fp4	94.91%	1
53	DeepSeek V3.1DeepSeek	—	$0.25	$0.95	164K	fp4	99.93%	1Open weights
54	DeepSeek V4 Flash 0423DeepSeek	—	$0.09	$0.18	1M	fp8	99.73%	1Open weights
55	DeepSeek V4 Flash 0731DeepSeek	—	$0.06	$0.18	1M	fp8	100.0%	1Open weights
56	DeepSeek V4 Pro 0423DeepSeek	—	$1.30	$2.60	1M	fp8	100.0%	1Open weights
57	GLM 4.6Z.ai	—	$0.50	$2.00	203K	fp4	100.0%	1Open weights
58	GLM 4.7Z.ai	—	$0.40	$1.75	203K	fp4	81.77%	1Open weights
59	GLM 5.2Z.ai	—	$0.563	$1.80	1M	fp4	99.55%	1Open weights
60	Hy3Tencent	—	$0.13	$0.53	262K	fp4	100.0%	1Open weights
61	Kimi K2.6Moonshot AI	—	$0.75	$3.50	262K	fp4	99.01%	1
62	Llama 3.1 70B InstructMeta	—	$0.40	$0.40	131K	fp8	71.99%	1
63	Mistral Small 3Mistral AI	—	$0.05	$0.08	33K	fp8	100.0%	1Open weights
64	Qwen3.5-27BQwen	—	$0.26	$2.60	262K	fp8	100.0%	1Open weights
65	Qwen3.5-35B-A3BQwen	—	$0.14	$1.00	262K	fp8	100.0%	1Open weights
66	Qwen3.6 27BQwen	—	$0.32	$3.20	262K	fp8	98.79%	1Open weights

DeepInfra on ModelCap: common questions

How many AI models does DeepInfra serve?

66 tracked models as of 1 October 2026: 66 language models (45 with a public ModelCap Index position), across 68 live endpoints observed through OpenRouter.

What is the best model available on DeepInfra?

MiMo-V2.6-Pro from Xiaomi is the highest-ranked language model DeepInfra serves, at ModelCap Index position #6 and $0.87 per 1M output tokens on DeepInfra. The table lists every model in Index order.

What is the cheapest model on DeepInfra?

Mistral Nemo at $0.03 per 1M output tokens ($0.019 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?

MiMo-V2.6-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 98.52% 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 1 October 2026. Publication and page updates happen separately. Offers retained from a failed refresh are marked stale in the data and are not counted here.