Open Source · framework

Use DeepSeek V2.5 with LlamaIndex

Updated 2026-05-21 · By XALEN

How to use DeepSeek V2.5 (Open Source, 236B (21B active)) with LlamaIndex. Install, authenticate, and make your first API call in minutes. Working code example included.

Model

DeepSeek V2.5

236B (21B active) · 128K context · $0.04 input

Framework

LlamaIndex

framework · pip install llama-index-llms-openai-like

1. Install

pip install llama-index-llms-openai-like

2. Code

from llama_index.llms.openai_like import OpenAILike

llm = OpenAILike(
    model="deepseek-v2-5",
    api_key="xln_test_YOUR_KEY",
    api_base="https://api.xalen.io/v1"
)

response = llm.complete("Explain DeepSeek V2.5")
print(response.text)

DeepSeek V2.5 with Other Frameworks

Python JavaScript React Next.js Flask FastAPI Django Express.js

Other Models with LlamaIndex

Vedika Standard Vedika Fast Vedika Pro Ultra Vedika Vision Vedika Code Vedika Astro Agent

200+ models. One API. Works with any framework.

Get API Key

Last updated: 2026-05-21