---
title: "LLM Models We Integrate: Kensink Labs"
description: "Direct integration with the models that matter: OpenAI GPT, Anthropic Claude, Google Gemini, Llama, Mistral, embeddings, and speech-to-text."
source: "https://www.kensink.com/models/"
canonical: "https://www.kensink.com/models/"
---
LLM MODELS · DIRECT INTEGRATION

# We integrate the model like we integrate Postgres.

Direct API integration against frontier and open-weight models. No LangChain, no orchestration vendor, no migration every six months. A thin abstraction so swapping models is a config change.

[Start a conversation →](https://www.kensink.com/contact) [How we work →](https://www.kensink.com/expertise)

[

01

## Claude

Direct Claude integration, eval-tested, no framework lock-in.

Capability Read more

](https://www.kensink.com/models/claude)[

![Kimi by Moonshot AI](https://www.kensink.com/assets/logos/kimi-icon.svg)02

## Kimi

Latest: Kimi K2.7. Open-weight frontier coding at a fraction of closed-leader cost, hosted or self-hosted.

Capability Read more

](https://www.kensink.com/models/kimi)[

![Sakana Fugu](https://www.kensink.com/assets/logos/fugu-icon.svg)03

## Sakana Fugu

Fugu and Fugu Ultra: a multi-agent system delivered as one model. Frontier results, no single-vendor dependency.

Capability Read more

](https://www.kensink.com/models/fugu)[

![OpenAI](https://www.kensink.com/assets/logos/openai-icon.svg)04

## OpenAI GPT

Production GPT integration with evals and cost control.

Capability Read more

](https://www.kensink.com/models/openai-gpt)[

05

## Google Gemini

Large context and multimodal understanding, integrated directly.

Capability Read more

](https://www.kensink.com/models/gemini)[

06

## Llama

Open-weight models you host yourself, for control and residency.

Capability Read more

](https://www.kensink.com/models/llama)[

07

## Mistral

Compact, efficient models, hosted or self-run.

Capability Read more

](https://www.kensink.com/models/mistral)[

08

## DeepSeek

Open-weight reasoning models, evaluated against your tasks.

Capability Read more

](https://www.kensink.com/models/deepseek)[

09

## Embeddings

Vector representations powering RAG, search, and recommendations.

Capability Read more

](https://www.kensink.com/models/embeddings)[

10

## Whisper & Speech-to-Text

Accurate transcription wired into your LLM pipeline.

Capability Read more

](https://www.kensink.com/models/whisper-speech)

Share[](https://twitter.com/intent/tweet?url=https%3A%2F%2Fwww.kensink.com%2Fmodels%2F&text=LLM%20Models%20%C2%B7%20Kensink%20Labs)[](https://www.linkedin.com/sharing/share-offsite/?url=https%3A%2F%2Fwww.kensink.com%2Fmodels%2F)

[View .md](https://www.kensink.com/models.md)

\[EXPLORE THE STACK\]

[

Languages

Languages & Frameworks

Browse](https://www.kensink.com/languages)[

Technologies

Technologies & Infrastructure

Browse](https://www.kensink.com/technologies)[

Design Patterns

Application Design Patterns

Browse](https://www.kensink.com/patterns)[

Project Types

Project Types

Browse](https://www.kensink.com/services)

SENIOR ENGINEERS · APPLIED K

## Not sure which fits?  
Tell us the problem.

We will tell you honestly what to build with, and what to skip. Senior engineers, eval-gated, full source ownership. Sized to the work.

[Book a 15-min intro →](https://www.kensink.com/contact)