OVH AI Endpoints

Verified ✅ Truly Free

OVHcloud's AI Endpoints in Beta. Access open source models hosted in Europe including Qwen3Guard, Audio, and Image generation models.

Free Quotas Beta European Hosting Community Pick OpenAI Compatible
Get API Key Suggest Edit
7

Overview

Provider Type

API

API Endpoint

https://oai.endpoints.kepler.ai.cloud.ovh.net/v1

Free Tier Highlights

2 RPM (Anonymous) / 400 RPM (Auth)

Why Choose OVH AI Endpoints?

OVHcloud's AI Endpoints serve open models hosted in Europe, including GPT-OSS 120B, Llama 3.3 70B, Mistral Small 3.2, Qwen2.5 VL, Whisper and BGE-M3 embeddings, and some can be called anonymously.

Best for: EU-hosted prototypes and GDPR-minded projects.

Watch out: free anonymous use is limited to 2 requests/minute per IP per model; higher limits are pay-as-you-go.

Quick Start Guide

1

Visit OVH AI Endpoints

2

Sign up / Login

3

Generate Token (for 400 RPM limit)

4

Use endpoint URL

Available Models

26 free models
Model Name ID Context Capabilities
GPT OSS 120B Free
gpt-oss-120b
128 000 Context
-
BGE-M3 (embeddings) Free
bge-m3
8 000 Context
-
Whisper Large v3 Turbo Free
whisper-large-v3-turbo
- -
Qwen2.5 VL 72B Instruct Free
Qwen2.5-VL-72B-Instruct
32 000 Context
-
Mistral Small 3.2 24B Free
Mistral-Small-3.2-24B-Instruct-2506
128 000 Context
-
Llama 3.3 70B Instruct Free
Meta-Llama-3_3-70B-Instruct
128 000 Context
-
GPT OSS 20B Free
gpt-oss-20b
128 000 Context
-
Qwen3.5 9B Free
Qwen3.5-9B
Long context
-
Qwen3.5 397B A17B Free
Qwen3.5-397B-A17B
Long context
-
Qwen3.6 27B Free
Qwen3.6-27B
Long context
-
Qwen3.8 27B Free
Qwen3.8-27B
Long context
-
Qwen3Guard-Gen-0.6B (Beta) Free
Qwen3Guard-Gen-0.6B
32 000 tokens
Text
Qwen3Guard-Gen-8B (Beta) Free
Qwen3Guard-Gen-8B
32 000 tokens
Text
stable-diffusion-xl-base-v10 Free
stable-diffusion-xl-base-v10
3.5B Params
Vision
nvr-tts-es-es Free
nvr-tts-es-es
Audio
Audio
nvr-tts-de-de Free
nvr-tts-de-de
Audio
Audio
nvr-tts-en-us Free
nvr-tts-en-us
Audio
Audio
nvr-tts-it-it Free
nvr-tts-it-it
Audio
Audio
DeepSeek V4.1 Flash 1017 Free
DeepSeek-V4.1-Flash-1017
1 050 000 Context
-
Qwen3.6 35B A3B Free
qwen3.6-35b-a3b
262 000 Context
-
Gemma 4 26B A4B Free
gemma-4-26b-a4b
262 000 Context
-
Qwen3 VL 30B A3B Free
qwen3-vl-30b-a3b
33 000 Context
-
Nemotron 3 Nano 30B Free
nemotron3-nano-30b
262 000 Context
-
Nemotron 3 Nano Omni 30B Free
nemotron3-omni-30b
131 000 Context
-
Laguna XS 2.1 Free
laguna-xs-2.1
131 000 Context
-
Solaria 3 Free
solaria-3
- -

Integration Examples

Ready-to-use code snippets for your applications.

main.py
import os
from openai import OpenAI

# Set your API key
# export OVH_AI_ENDPOINTS_ACCESS_TOKEN=<your-access-token>

client = OpenAI(
    base_url="https://oai.endpoints.kepler.ai.cloud.ovh.net/v1",
    api_key=os.getenv("OVH_AI_ENDPOINTS_ACCESS_TOKEN", "empty-if-anonymous")
)

response = client.chat.completions.create(
    model="gpt-oss-120b",
    messages=[
        {"role": "user", "content": "Hello!"}
    ],
    temperature=0,
    max_tokens=1024
)

print(response.choices[0].message.content)

Free Tier Pricing & Limits

Rate Limit

Requests per minute

2 Requests per minute per IP per model (anonymousfree); 400 Requests per minute with a key (pay-as-you-go)

Daily Quota

Requests per day

Unspecified

Token Limit

Tokens per minute

Unspecified

Monthly Quota

Per month limit

Beta Access

Limitations & Considerations

Beta service, may end or change

2 requests/minute for anonymous usage

Requires token for higher limits (400 RPM)

Community Hub

Live

Join the discussion, share tips, and rate OVH AI Endpoints.

Quick Reactions

Add Discussion

Comments are moderated. Be helpful and respectful.

Recent Activity

0 comments

Ready to Get Started?

Join thousands of developers using OVH AI Endpoints

Start Building Now