DeepInfra

Serverless Featured

Visit Provider Docs Pricing

Serverless inference for open-source LLMs and generative models. Pay-per-token with fast cold starts.

Features

OpenAI Compatible

Streaming

Batching

Fine-tuning

Embeddings

Vision

Audio

Function Calling

JSON Mode

Compliance

SOC 2

HIPAA

GDPR

Pricing Model

per token

Free tier: Available

Details

Models 82

API Base https://api.deepinfra.com/v1/openai

Audio / MusicImage GenerationLLM Inference

Model Catalog (82)

Model	Type	Input $/1M	Output $/1M	Context	Speed
BGE-M3 BAAI	embedding	$0.010	—	—	—
Claude 4.5 Haiku Anthropic	llm	$1.00	$5.00	200k	—
Claude 4.6 Sonnet Anthropic	llm	$3.00	$15.00	1M	—
Claude Fable 5 Anthropic	llm	$10.00	$50.00	1M	—
Claude Opus 4.7 Anthropic	llm	$5.00	$25.00	1M	—
Claude Opus 4.8 Anthropic	llm	$5.00	$25.00	1M	—
Claude Opus 5 Anthropic	llm	$5.00	$25.00	1M	—
Claude Opus 5.5 Anthropic	llm	$4.00	$20.00	1M	—
Claude Sonnet 5 Anthropic	llm	$3.00	$15.00	1M	—
Claude Sonnet 5.5 Anthropic	llm	$2.00	$10.00	1M	—
DeepSeek V3 DeepSeek · 671B MoE	llm	$0.320	$0.890	164k	—
DeepSeek V3.1 DeepSeek · 671B MoE	llm	$0.250	$0.950	164k	—
DeepSeek V3.2 DeepSeek · 671B MoE (37B active)	llm	$0.260	$0.380	—	—
DeepSeek V4 Flash DeepSeek	llm	$0.090	$0.180	—	—
DeepSeek V4 Flash DeepSeek 0731	llm	$0.060	$0.180	1.0M	—
DeepSeek V4 Pro DeepSeek	llm	$1.30	$2.60	—	—
DeepSeek V4 Pro DeepSeek 0813	llm	$1.30	$2.60	1.0M	—
DeepSeek V4.1 Flash DeepSeek	llm	$0.200	$0.600	1.0M	—
Flux 2 Max Black Forest Labs	image gen	$0.070/img	—	—	—
GLM 4.6 Zhipu AI · 355B	llm	$0.500	$2.00	203k	—
GLM 4.7 Zhipu AI	llm	$0.060	$0.400	—	—
GLM 5 Zhipu AI · 744B	llm	$0.600	$2.08	—	—
GLM 5.2 Z.ai	llm	$0.750	$2.40	1.0M	—
GLM 5.3 Z.ai	llm	$0.900	$4.00	1.0M	—
GLM 5.3 Flash Z.ai	llm	$0.150	$0.500	1.0M	—
GPT OSS 120B OpenAI · 117B MoE (5.1B active)	llm	$0.037	$0.170	131k	—
Gemini 2.5 Flash Google	llm	$0.300	$2.50	1M	—
Gemini 2.5 Pro Google	llm	$1.25	$10.00	1M	—
Gemini 3 Pro Image Google	image gen	$120.00	—	—	—
Gemini 3.1 Flash Lite Google	llm	$0.250	$1.50	1M	—
Gemini 3.1 Pro Google	llm	$2.00	$12.00	1M	—
Gemini 3.5 Flash Google	llm	$1.50	$9.00	1M	—
Gemini 3.7 Flash Google	llm	$0.750	$3.75	1M	—
Gemma 3 12B Google · 12B	llm	$0.050	$0.150	131k	—
Gemma 3 27B Google · 27B	llm	$0.080	$0.160	131k	—
Gemma 3 4B Google · 4B	llm	$0.050	$0.100	131k	—
Gemma 4 26B A4B Google	llm	$0.070	$0.340	262k	—
Gemma 4 31B Google · 27B	llm	$0.150	$0.400	262k	—
Granite 4.2 8B IBM	llm	$0.060	$0.250	131k	—
Hy4 Preview Tencent	llm	$0.834	$2.50	1.0M	—
Inkling Thinking Machines	llm	$0.950	$4.05	524k	—
Inkling Small Thinking Machines	llm	$0.450	$1.20	524k	—
Kimi K2.5 Moonshot AI · 1T MoE (32B active)	llm	$0.450	$2.25	—	—
Kimi K2.6 Moonshot AI · 1T MoE (32B active)	llm	$0.750	$3.50	262k	—
Kimi K2.7 Code Moonshot AI · 1T MoE (32B active)	code	$0.680	$3.40	262k	—
Kimi K3 Moonshot AI	llm	$2.85	$14.25	1.0M	—
MiMo V2.6 Flash Xiaomi	llm	$0.140	$0.280	1.0M	—
MiMo V2.6 Pro Xiaomi	llm	$0.430	$0.870	1.0M	—
MiniMax M2.5 MiniMax	llm	$0.150	$1.15	—	—
MiniMax M3 MiniMax	llm	$0.280	$1.10	524k	—
Muse Glimmer 30B Meta	llm	$0.300	$1.20	131k	—
Qwen 3 235B Alibaba · 235B MoE	llm	$0.090	$0.550	262k	—
Qwen 3 32B Alibaba · 32B	llm	$0.080	$0.280	41k	—
Qwen 3 Max Alibaba	llm	$1.20	$6.00	—	—
Qwen 3 Max Thinking Alibaba	llm	$1.20	$6.00	—	—
Qwen 3 TTS Alibaba	text to_speech	$0.00002/character	see notes	—	—
Qwen 3.5 122B Alibaba · 122B MoE (10B active)	llm	$0.290	$2.40	—	—
Qwen 3.5 35B Alibaba · 35B MoE (3B active)	llm	$0.140	$1.00	—	—
Qwen 3.5 397B Alibaba · 397B MoE (17B active)	llm	$0.450	$3.00	—	—
Qwen 3.5 72B Alibaba · 72B	llm	$0.260	$2.60	—	—
Qwen 3.5 9B Alibaba · 9B	llm	$0.100	$0.150	—	—
Qwen3 14B Alibaba	llm	$0.120	$0.240	41k	—
Qwen3 30B A3B Alibaba	llm	$0.120	$0.500	41k	—
Qwen3 Embedding 0.6B Alibaba · 0.6B	embedding	$0.010	—	—	—
Qwen3 Embedding 4B Alibaba · 4B	embedding	$0.020	—	—	—
Qwen3 Embedding 8B Alibaba · 8B	embedding	$0.010	—	—	—
Qwen3 Next 80B A3B Instruct Alibaba	llm	$0.090	$1.10	262k	—
Qwen3 VL 235B A22B Instruct Alibaba	multimodal	$0.200	$0.880	262k	—
Qwen3 VL 30B A3B Instruct Alibaba	multimodal	$0.150	$0.600	262k	—
Qwen3.6 27B Alibaba	llm	$0.320	$3.20	262k	—
Qwen3.6 35B A3B Alibaba	llm	$0.100	$0.950	262k	—
Qwen3.7 Max Alibaba	llm	$2.50	$7.50	—	—
Qwen3.8 2.4T A95B Alibaba	llm	$2.00	$6.00	262k	—
Qwen3.8 27B Alibaba	llm	$0.200	$2.50	262k	—
Qwen3.8 Flash Alibaba	llm	$0.113	$0.382	1M	—
Qwen3.8 Max Alibaba	llm	$1.65	$4.95	256k	—
Seed 2.0 Code ByteDance	llm	$0.500	$3.00	256k	—
Seed-2.0-Mini ByteDance	llm	$0.100	$0.400	256k	—
Seedance 1.5 Pro ByteDance	video gen	$1.20	—	—	—
Seedance 2.0 ByteDance	video gen	$4.70	—	—	—
Veo 3.1 Google	video gen	$0.400/sec	—	—	—
gpt-oss-20b OpenAI	llm	$0.030	$0.140	131k	—

DeepInfra API FAQ

Is the DeepInfra API free?

Yes, DeepInfra has a free tier. Paid usage is billed per model.

What is the cheapest DeepInfra model?

BGE-M3 at $0.01 in per 1M tokens, based on the 82 models we track for DeepInfra.

How many models does DeepInfra offer?

We track 82 DeepInfra models: 69 LLM, 4 embedding, 3 video generation, 2 image generation, 2 multimodal, 1 coding, 1 text-to-speech.

Is the DeepInfra API OpenAI-compatible?

Yes. DeepInfra exposes an OpenAI-compatible endpoint, so you can use the OpenAI SDKs by changing the base URL and API key.