OVHcloud AI Endpoints
Inference cloud
- Catalog status
- List checked 24m ago
- Last successful check
- 2026-11-20 (24m ago)
- Latest attempt
- 24m ago: succeeded
"List checked" means we fetched the provider's model list in the last 3 hours. That shows the list is reachable and current, not that inference works.
Free quota
Free
AI Endpoints sandbox
- Kind
- Recurring
- Credit
- Not documented
- Tokens granted
- Not documented
- Expires after
- Not documented
- Renews every
- Not documented
- Card required
- No
- Phone required
- No
- Trains on prompts
- No
- Commercial use
- Not documented
- Eligibility
- Not documented
- Note
- Free access is keyless: 2 RPM per IP, per model. API keys get 400 RPM per project, but need a payment method and are billed by token, so 400 RPM isn't free. Embedding models are served on /v1/embeddings, not /chat/completions. Qwen3Guard models are served on /v1/chat/completions.
- Source
- Provider page, checked 2026-11-01
| Limits | Per minute | Per day | Per month |
|---|---|---|---|
| Requests | 2 | Not documented | Not documented |
| Tokens | Not documented | Not documented | Not documented |
Models covered
Qwen3.5-397B-A17BQwen3.6-27Bgpt-oss-120bgpt-oss-20bMeta-Llama-3_3-70B-InstructMistral-Small-3.2-24B-Instruct-2506Qwen3.5-9BQwen2.5-VL-72B-InstructQwen3-Embedding-8Bbge-m3bge-multilingual-gemma2Qwen3Guard-Gen-8BQwen3Guard-Gen-0.6BQwen3.8-27B
How to call it
- OpenAI-compatible
- Yes, Chat Completions at the base URL
- Base URL
https://oai.endpoints.kepler.ai.cloud.ovh.net/v1
curl "https://oai.endpoints.kepler.ai.cloud.ovh.net/v1/chat/completions" \
-H "Authorization: Bearer $OVH_AI_ENDPOINTS_ACCESS_TOKEN" \
-H "Content-Type: application/json" \
-d '{"model": "DeepSeek-V4.1-Flash-1017", "messages": [{"role": "user", "content": "Hello"}]}'
Models
| Model ID | Comparison key | Input $/1M | Output $/1M | Price checked |
|---|---|---|---|---|
DeepSeek-V4.1-Flash-1017 |
unknown | unknown | — | |
Meta-Llama-3_3-70B-Instruct |
unknown | unknown | — | |
Mistral-Small-3.2-24B-Instruct-2506 |
unknown | unknown | — | |
Olala-v1.0-mini-7A1B |
unknown | unknown | — | |
Olala-v1.0-summarise |
unknown | unknown | — | |
Olala-v1.0-translate |
unknown | unknown | — | |
Qwen2.5-VL-72B-Instruct |
unknown | unknown | — | |
Qwen3-Embedding-8B |
unknown | unknown | — | |
Qwen3.5-397B-A17B |
unknown | unknown | — | |
Qwen3.5-9B |
unknown | unknown | — | |
Qwen3.6-27B |
unknown | unknown | — | |
Qwen3.8-27B |
unknown | unknown | — | |
Qwen3Guard-Gen-0.6B |
unknown | unknown | — | |
Qwen3Guard-Gen-8B |
unknown | unknown | — | |
bge-m3 |
unknown | unknown | — | |
bge-multilingual-gemma2 |
unknown | unknown | — | |
clef-flash |
unknown | unknown | — | |
gemma-4-26b-a4b |
unknown | unknown | — | |
gpt-oss-120b |
openai/gpt-oss-120b |
$0.09 | $0.47 | models.dev, synced 13h ago |
gpt-oss-20b |
openai/gpt-oss-20b |
$0.05 | $0.18 | models.dev, synced 13h ago |
laguna-xs-2.1 |
unknown | unknown | — | |
nemotron3-nano-30b |
unknown | unknown | — | |
nemotron3-omni-30b |
unknown | unknown | — | |
nvr-tts-de-de |
unknown | unknown | — | |
nvr-tts-en-us |
unknown | unknown | — | |
nvr-tts-es-es |
unknown | unknown | — | |
nvr-tts-it-it |
unknown | unknown | — | |
qwen3-vl-30b-a3b |
unknown | unknown | — | |
qwen3.6-35b-a3b |
unknown | unknown | — | |
solaria-3 |
unknown | unknown | — | |
stable-diffusion-xl-base-v10 |
unknown | unknown | — | |
whisper-large-v3 |
unknown | unknown | — | |
whisper-large-v3-turbo |
unknown | unknown | — |