# Model Providers

Entry Point AI allows you to connect one or more [model providers](/key-concepts/model-providers.md) to run prompts, fine-tune models, and host models for inference. Our software acts as an abstraction layer that makes it easy to switch between or compare models from different providers.

<figure><img src="/files/73QKIsS2POvaep8GI7Mf" alt=""><figcaption></figcaption></figure>

We support both proprietary model providers (OpenAI, Anthropic, Google AI), and open-source model providers (Replicate, Groq).

Model providers can be connected to Entry Point from the [Integrations](https://app.entrypointai.com/integrations) page found in the top nav bar. While the process of setting up these integrations can vary slightly from provider to provider, most of them simply require that you input an API key to get going.&#x20;

{% hint style="danger" %}
If you disconnect a model provider, the associated models will become inaccessible through Entry Point.&#x20;
{% endhint %}

Here are some of our top model providers:

### Replicate

Replicate is a leading provider of open-source models.

| Models for Inference                                                                          | Models for Fine-tuning                                                                     |
| --------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------ |
| <p>Mistral 7B (chat and pretrained versions)</p><p>Llama 2 (chat and pretrained versions)</p> | <p>Mistral 7B (chat and pretrained versions)<br>Llama 2 (chat and pretrained versions)</p> |

### OpenAI

OpenAI offers the GPT family of models for inference and fine-tuning. If you have early access to other models that you'd like to use inside of Entry Point, shoot us an email at <support@entrypointai.com>

| Models for Inference                                                   | Models for Fine-tuning                         |
| ---------------------------------------------------------------------- | ---------------------------------------------- |
| <p>GPT-4o<br>GPT-4 Turbo<br>GPT-4 (Original, 32k)<br>GPT-3.5 Turbo</p> | <p>GPT-3.5 Turbo<br>Davinci 2<br>Babbage 2</p> |

### Groq Cloud

Groq is the fast inference engine for AI.

| Models for Inference                                          | Models for Fine-tuning |
| ------------------------------------------------------------- | ---------------------- |
| <p>Mixtral 8x7b</p><p>Llama 3 (8B and 70B)</p><p>Gemma 7B</p> | N/A                    |

### AI21

AI21 provides the Jurassic-2 family of models.

| Models for Inference                                          | Models for Fine-tuning |
| ------------------------------------------------------------- | ---------------------- |
| <p>Jurassic-2 Light<br>Jurassic-2 Mid<br>Jurassic-2 Ultra</p> | N/A                    |

{% hint style="warning" %}
AI21 has paused the availability of their fine-tuning APIs under further notice.
{% endhint %}

### Google AI

Google AI provides the Gemini family of models for inference.&#x20;

| Models for Inference | Models for Fine-tuning |
| -------------------- | ---------------------- |
| Gemini 1.0 Pro       | N/A                    |

{% hint style="info" %}
Google AI offers a fine-tuning API but its not integrated with Entry Point yet.
{% endhint %}

### Anthropic

Anthropic provides the Claude family of models for inference.&#x20;

| Models for Inference                                            | Models for Fine-tuning |
| --------------------------------------------------------------- | ---------------------- |
| <p>Claude 3 Opus</p><p>Claude 3 Sonnet</p><p>Claude 3 Haiku</p> | N/A                    |


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.entrypointai.com/key-concepts/model-providers.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.