Updated

DeepInfra

Inference cloud serving hundreds of open-weight models — chat, embeddings, speech, image and video — through an OpenAI-compatible API, plus dedicated GPUs for your own models.

7 open-source projects declare DeepInfra in their code; 2 makers' products we track use it.

No free tierPaid from Pay per token

Profile checked

Ask your AI about this, with this page as the source:ChatGPT ↗Claude ↗Perplexity ↗
Recent changesFollow DeepInfra (RSS)
  • Oct 2026 Now covered: DeepInfra, for llm api.

How makers use DeepInfra

Makers' products we track use DeepInfra for LLM API (2).

What DeepInfra is used for

LLM API · Open models, hosted →Low per-token prices on a large catalog of open-weight models, plus embeddings, speech and image models, behind an OpenAI-compatible API.Trade-off: No free tier — a card or prepaid balance comes first — and a default cap of 200 concurrent requests per model until you ask for more.
Open Model Hosting & GPUs · Serverless model API →Hundreds of open models — chat, embeddings, speech, image and video — through one OpenAI-compatible API, plus dedicated GPUs for your own models.Trade-off: Hosted only; dedicated GPUs bill per GPU-hour while they run.
Hugging Face77Modal60Baseten23Runpod19Together AI12DeepInfra0Makers' products on each, for open model hosting & GPUs

DeepInfra vs the alternatives makers chose

For LLM API, 2 of the 923 makers' products we know the choice of picked DeepInfra (0%), behind OpenAI API (80%).

OptionMakers' products2026 launchesFree forAt 10,000 usersOpen source
OpenAI API736 80%77%~1,000 users$10/mo · GPT-6 LunaNo
Claude252 27%21%~1,000 users$10/mo · Claude Haiku 5.5No
Gemini API99 11%12%~360 users$28/mo · Gemini 3.1 Flash-LiteNo
Groq39 4%3%~1,500 users$6.75/mo · GPT OSS 20BNo
Azure OpenAI22 2%0%——No
Mistral AI15 2%1%~1,700 users$6.00/mo · Ministral 3 (3B)No
DeepInfra2 0%0%~1,600 users—Unknown

For LLM API: how many of the 923 makers' products we know the LLM API of chose each option, and the share among the 256 launched in 2026, since a tool that came first has products from every year, and what each costs at 10,000 users.

Costs are the cheapest plan that fits, from each vendor's pricing page, at the usage our stack planner assumes for a product with that many users. Free for: the users its cheapest plan a business can use stays free for.

When to use DeepInfra, and when not

Something else fits better if

  • You want the strongest general models and the widest ecosystem of examples and integrations: OpenAI API or Claude or Gemini API.
  • Token cost dominates your budget, for example high-volume batch or agent workloads: DeepSeek or Groq or Mistral AI.
  • You want to try or mix many models with one key and one bill, and fall back when a provider is down: OpenRouter.
  • You want that routing and spend control but self-hosted, with keys and logs on your servers: LiteLLM.
  • The open model you need is already served by a host and you want to pay per token: Together AI or Fireworks AI or Hugging Face.
From each decision's guide, written from what makers' products chose and why.

DeepInfra is an inference cloud that hosts other labs' open-weight models — DeepSeek, Qwen, Kimi, Llama, Gemma, Nemotron and many more — and bills per token, with no servers to run. Besides chat models it serves embeddings and rerankers, speech recognition and text-to-speech, and image and video generation, and it also lists a few closed models such as Anthropic's and Google's.

More about DeepInfra

You call it with the OpenAI SDK by pointing the base URL at https://api.deepinfra.com/v1/openai and passing a DeepInfra key; an Anthropic Messages endpoint at https://api.deepinfra.com/anthropic lets the Anthropic SDK and Claude Code use its models too. There are also official deepinfra packages for Node and Python and a Vercel AI SDK provider.

Useful for a small team: one account covers chat, embeddings and speech; prompt caching is automatic and bills repeated prefixes at a lower cached-input rate; a per-request service tier trades price for speed (priority at 1.5× the base price, flex at 0.8× for non-urgent work), and the Batch API runs asynchronous jobs at 20% off. Inputs and outputs are held in memory only and not used for training, and when you outgrow shared capacity you can deploy your own model or LoRA on dedicated GPUs billed by the minute.

The main limits: there is no free tier — you must add a card or prepay before the first call — and each account starts at 200 concurrent requests per model until you ask for more. You only get the models DeepInfra chooses to host, and for the closed models it resells, the model maker's own data and training terms apply.

What Hacker News says about DeepInfra

On Hacker News, 52% of comments about DeepInfra recommend it and 0% criticize it: net praise +52%.

By year

YearComments choosing it, per monthNet praise
20230.1—
20240.1—
20250.3+60%
2026 (8 mo)1.3+54%

In their words

  • “I also switched to DeepInfra ... who serves an fp4 version admittedly, but I've had no issues with it” HN, Aug 2026
  • “I am quite happy with them so far.” HN, Aug 2026

Hacker News comments January 2023 to August 2026 that name DeepInfra in a sentence like "I use" or "we moved to", Claude read a sample of 22 and the counts scale it to all of them; net praise is comments recommending it minus comments criticizing it, as a share of all comments about it. How it's counted, and how HN compares with makers.

Who uses it

2 makers' products, each linked to the source that shows it, and 7 open-source projects that declare it in their code. The best known on Product Hunt: Triqai (105 followers).

Subprocessor list 2Declared in code 7How evidence is collected →

Open source: a project that declares DeepInfra as a dependency in its public code — verifiable, but not necessarily a live product.

Alternatives to DeepInfra

All alternatives by situation →

Further reading

Questions makers ask about DeepInfra

What is DeepInfra used for?

Inference cloud serving hundreds of open-weight models — chat, embeddings, speech, image and video — through an OpenAI-compatible API, plus dedicated GPUs for your own models. Makers' products we track use it for LLM API (2).

What are the alternatives to DeepInfra?

For LLM API, the other options makers' products we track chose most are OpenAI API (736), Claude (252), Gemini API (99), Groq (39), Azure OpenAI (22), Mistral AI (15). Each alternative, with the situation it fits, is on the DeepInfra alternatives page.

Can I use the OpenAI SDK?

Yes. Set the base URL to https://api.deepinfra.com/v1/openai, use a DeepInfra API token as the key, and pass a DeepInfra model name such as deepseek-ai/DeepSeek-V4-Flash. source ↗

Does it work with the Anthropic SDK or Claude Code?

Yes. DeepInfra exposes an Anthropic Messages-compatible endpoint at https://api.deepinfra.com/anthropic; set it as the SDK's base URL, or as ANTHROPIC_BASE_URL (without /v1) for Claude Code. source ↗

Does DeepInfra store my prompts or train on them?

DeepInfra says inputs are only held in memory during inference and outputs are deleted once returned, request content is not logged, and data is not used for training or shared — except when you call Google or Anthropic models, where those companies' policies apply. Batch jobs may be stored encrypted on disk for a short time. source ↗

What are the rate limits?

The default is 200 concurrent requests per model rather than a requests-per-minute cap, so throughput depends on how long each request takes. You can request an increase from the account dashboard. source ↗

Is there a free tier?

No. You have to add a card or prepay before using the API; prepaid accounts draw down a balance, invoiced accounts are billed monthly, and you can set a spending limit. source ↗

Can I pay less for work that isn't urgent?

Yes. The flex service tier costs 0.8× the base price in exchange for slower responses and occasional unavailability, and the Batch API returns results within 24 hours at 20% below real-time pricing. source ↗

Can I run my own fine-tuned model?

Yes. You can deploy a custom LLM or LoRA adapter on dedicated A100, H100, H200, B200 or B300 GPUs with autoscaling and an OpenAI-compatible endpoint, billed per GPU-hour at minute granularity. source ↗

Is DeepInfra free?

No permanent free tier; paid use starts at Pay per token. source ↗

Can AI coding agents work with DeepInfra?

No llms.txt, official MCP server or CLI found yet.

Who uses DeepInfra?

2 makers' products we track, each with a source, and 7 open-source projects declare it in their code. The best known of them on Product Hunt: Triqai (105 followers). source ↗