deepinfra base url

Not specified by the original author.

TL;DR

Pass models as HuggingFace-style ids (deepseek-ai/DeepSeek-V4-Flash, meta-llama/Meta-Llama-3.1-8B-Instruct). Set your baseURL to https://api.deepinfra.com/v1/openai exactly, with openai after v1.

Steps

1. Set your baseURL to https://api.deepinfra.com/v1/openai exactly, with openai after v1. Using https://api.deepinfra.com/v1 is the common failure that breaks every request. Pass models as HuggingFace-style ids (deepseek-ai/DeepSeek-V4-Flash, meta-llama/Meta-Llama-3.1-8B-Instruct). For cost tracking, read usage.estimated_cost off each response instead of estimating from your own token counts.

When to use

You are seeing this: Using https://api.deepinfra.com/v1 is the common failure that breaks every request. Use this skill when you run into "DeepInfra base URL ends in slash v1 slash openai, not slash v1".

When not to use

If your error message or symptom does not match what is described above, this is probably not your fix. Search for your exact error text instead of forcing this one to fit.

Versions

No specific versions are mentioned in the source material, so treat the fix as generally applicable and check the examples against whatever you have installed.

Why this happens

The original report does not dig into a root cause. It documents the symptom and the fix that resolved it.

Sources

• https://api.deepinfra.com/v1/openai
• https://api.deepinfra.com/v1

ai-lcr provider doc for DeepInfra (github.com/ai-lcr/ai-lcr website/content/docs/providers/deepinfra.mdx): 'The one quirk that trips everyone: DeepInfra serves the OpenAI-compatible API at /v1/openai/chat/completions, the /v1/ comes before openai.' Since createOpenAICompatible appends /chat/completions to the baseURL, the base must be https://api.deepinfra.com/v1/openai, not .../v1. Model ids are HuggingFace-style paths like deepseek-ai/DeepSeek-V4-Flash, and responses carry usage.estimatedcost for billing reconciliation. Resolved: Set your baseURL to https://api.deepinfra.com/v1/openai exactly, with openai after v1. Using https://api.deepinfra.com/v1 is the common failure that breaks every request. Pass models as HuggingFace-style ids (deepseek-ai/DeepSeek-V4-Flash, meta-llama/Meta-Llama-3.1-8B-Instruct). For cost tracking, read usage.estimatedcost off each response instead of estimating from your own token counts. Source: https://github.com/ai-lcr/ai-lcr/blob/HEAD/website/content/docs/providers/deepinfra.mdx

Not specified by the original author.

TL;DR

Set base_url to https://api.deepinfra.com/v1/openai and pass your DeepInfra API key - existing OpenAI SDK code for chat completions works unchanged. Use DeepInfra model IDs like meta-llama/Meta-Llama-3.1-70B-Instruct instead of OpenAI model names. DeepInfra exposes an OpenAI-compatible endpoint.

Fix

1. Set base_url to https://api.deepinfra.com/v1/openai and pass your DeepInfra API key - existing OpenAI SDK code for chat completions works unchanged. Expected: You get the expected result; the problem is gone.

2. Use DeepInfra model IDs like meta-llama/Meta-Llama-3.1-70B-Instruct instead of OpenAI model names. Expected: You get the expected result; the problem is gone.

When to use

• You are setting up or using this DeepInfra with the OpenAI SDK feature.
• The symptom matches: point base_url at api.deepinfra.com.

When NOT to use

• Unrelated DeepInfra with the OpenAI SDK issues (different feature, different failure).
• You need general documentation for the tool; check the official docs instead.

Compatibility

Reported against DeepInfra with the OpenAI SDK (mentions 3.1).

Variant phrasings

point base_url at api.deepinfra.com

Why it happens

No new SDK needed. Swap the base URL and keep your OpenAI client code.

Edge cases

• If your error message differs even slightly, this is probably a different issue; search the exact text.
• Behavior can change between releases; the linked source reflects the versions above.
• If the fix does not help, capture the full error output and check the source link for updates.

Source: No new SDK needed. Swap the base URL and keep your OpenAI client code.

Verified resolution: [DeepInfra docs]: DeepInfra exposes an OpenAI-compatible endpoint. Set base_url to https://api.deepinfra.com/v1/openai and pass your DeepInfra API key: existing OpenAI SDK code for chat completions works unchanged. Use DeepInfra model IDs like meta-llama/Meta-Llama-3.1-70B-Instruct instead of OpenAI model names. If you get a model-not-found error, the ID is wrong or the model is not served on DeepInfra: check the model list first.

Source URL: https://deepinfra.com/docs/openai_api

Not specified by the original author.

TL;DR: Pick the right endpoint family first. Point your OpenAI SDK at YOURHOST.deepinfra.com/v1/openai for chat, embeddings, and image gen. If you need speech recognition, object detection, or other model types with no OpenAI equivalent, use the native POST YOURHOST.deepinfra.com/v1/inference/{model_name} instead.

Fix

1. Pick the right endpoint family first. Expected: this specific failure stops.

2. Point your OpenAI SDK at YOUR_HOST.deepinfra.com/v1/openai for chat, embeddings, and image gen. Expected: this specific failure stops.

3. If you need speech recognition, object detection, or other model types with no OpenAI equivalent, use the native POST YOURHOST.deepinfra.com/v1/inference/{modelname} instead. Expected: the request routes instead of erroring.

Details

In both cases auth is a Bearer token in the Authorization header, available from the DeepInfra dashboard API keys page. Context: DeepInfra docs API Reference intro (deepinfra/docs api-reference/introduction.mdx): The DeepInfra API offers two endpoint families. OpenAI-compatible endpoints live at /v1/openai and are a drop-in replacement for OpenAI chat completions, embeddings, and image generation with any OpenAI SDK. DeepInfra native endpoints live at /v1/inference/{model_name} and expose model types the OpenAI schema does not, such as speech recognition, object detection, and image classification. All endpoints need a Bearer token in the Authorization header.

When to use

You hit exactly this: DeepInfra has two endpoint families, pick openai or native inference.

When not to use

A different error, or the same symptom in a different tool. This page only covers the failure above.

Compatibility

the tool named in the title (versions mentioned here: v1).

DeepInfra docs API Reference intro (deepinfra/docs api-reference/introduction.mdx): The DeepInfra API offers two endpoint families. OpenAI-compatible endpoints live at /v1/openai and are a drop-in replacement for OpenAI chat completions, embeddings, and image generation with any OpenAI SDK. DeepInfra native endpoints live at /v1/inference/{modelname} and expose model types the OpenAI schema does not, such as speech recognition, object detection, and image classification. All endpoints need a Bearer token in the Authorization header. Resolved: Pick the right endpoint family first. Point your OpenAI SDK at https://api.deepinfra.com/v1/openai for chat, embeddings, and image gen. If you need speech recognition, object detection, or other model types with no OpenAI equivalent, use the native POST https://api.deepinfra.com/v1/inference/{modelname} instead. In both cases auth is a Bearer token in the Authorization header, available from the DeepInfra dashboard API keys page. Source: https://github.com/deepinfra/docs/blob/HEAD/api-reference/introduction.mdx