[{"available":false,"category":"Visual LLM","category_v2":"Text Generation","created_at":"2025-07-16T08:47:07Z","description":"Building upon Mistral Small 3 (2501), Mistral Small 3.2 (2506) adds state-of-the-art vision understanding and enhances long context capabilities up to 128k tokens without compromising text performance.\nWith 24 billion parameters, this model achieves top-tier capabilities in both text and vision tasks.\nThis model is a minor update of Mistral-Small-3.1-24B-Instruct-2503, which is an instruction-finetuned version of: Mistral-Small-3.1-24B-Base-2503.\n","documentation_url":"https://mistral-small-3-2-24b-instruct-2506.endpoints.kepler.ai.cloud.ovh.net/doc/readme","gradio_url":"https://playground.endpoints.ai.ovh.net/Mistral-Small-3.2-24B-Instruct-2506","id":"mistral-small-3-2-24b-instruct-2506","isNew":false,"licence_url":"https://mistral-small-3-2-24b-instruct-2506.endpoints.kepler.ai.cloud.ovh.net/doc/license","metadata":{"aliases":["mistralai/Mistral-Small-3.2-24B-Instruct-2506","mistral-small-3.2-24b-instruct-2506"],"api_base_url":"https://mistral-small-3-2-24b-instruct-2506.endpoints.kepler.ai.cloud.ovh.net/api/openai_compat/","benchmark_results":[{"benchmark_name":"Wildbench v2","benchmark_score":65.33},{"benchmark_name":"Arena Hard v2","benchmark_score":43.1},{"benchmark_name":"IF (Internal; accuracy)","benchmark_score":84.78},{"benchmark_name":"Infinite Generations (Internal; Lower is better)","benchmark_score":1.29},{"benchmark_name":"MMLU","benchmark_score":80.5},{"benchmark_name":"MMLU Pro (5-shot CoT)","benchmark_score":69.06},{"benchmark_name":"MATH","benchmark_score":69.42},{"benchmark_name":"GPQA Main (5-shot CoT)","benchmark_score":44.22},{"benchmark_name":"GPQA Diamond (5-shot CoT )","benchmark_score":46.13},{"benchmark_name":"MBPP Plus - Pass@5","benchmark_score":78.33},{"benchmark_name":"HumanEval Plus - Pass@5","benchmark_score":92.9},{"benchmark_name":"SimpleQA (TotalAcc)","benchmark_score":12.1},{"benchmark_name":"MMMU","benchmark_score":62.5},{"benchmark_name":"Mathvista","benchmark_score":67.09},{"benchmark_name":"ChartQA","benchmark_score":87.4},{"benchmark_name":"DocVQA","benchmark_score":94.86},{"benchmark_name":"AI2D","benchmark_score":92.91}],"context":"128k","deployment_information":{"datacenters":[{"country_code":"FR"}]},"model_base_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/v1","model_specs":{"capabilities":{"batch":true,"fill_in_the_middle":false,"function_calling":true,"input_modality":["text","image"],"output_modality":["text"],"reasoning":false,"response_format":["raw_text","json_object","json_schema"],"streaming":true,"supported_sampling_parameters":["temperature","stop","top_p","seed"],"use_case_tag":["code_chat","code_completion","code_explanation","emotion_detection","entity_extraction","multi_language_coding","multilingual","paraphrasing","sentiment_analysis","summarization","text_generation","text_only_compatible","translation","vision"],"web_search":false},"model_parameters":{"parameters_type":"vlm","params":{"context_size_k":128,"max_images_per_prompt":10,"max_output_token_k":128,"supported_aspect_ratios":["*"]}},"nb_of_billion_parameters":24,"quantization":"bf16"},"playground_url":"https://playground.endpoints.ai.ovh.net/Mistral-Small-3.2-24B-Instruct-2506","publishing_information":{"endpoint_publication_date":"30/06/2025","knowledge_cutoff":"None","languages":["*"],"licence_name":"Apache 2.0","licence_url":"https://mistral-small-3-2-24b-instruct-2506.endpoints.kepler.ai.cloud.ovh.net/doc/license","model_eol_date":"04/11/2026","model_publication_date":"20/06/2025","publisher":"mistralai","series":"mistral","source":"huggingface","source_link":"https://huggingface.co/mistralai/Mistral-Small-3.2-24B-Instruct-2506"},"router_base_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/v1","tags":["Vision","Streaming"],"usage_information":{"avg_throughtput":2,"avg_throughtput_unit":"empirical","pricing":[{"price":0.09,"price_unit":"million_input_tokens"},{"price":0.28,"price_unit":"million_output_tokens"}],"rate_limit":{"rate_limit_api":[{"rate_limit_unit":"req/m","rate_limit_value":400}],"rate_limit_api_anonymous":[{"rate_limit_unit":"req/m","rate_limit_value":2}],"rate_limit_playground":[{"rate_limit_unit":"req/m","rate_limit_value":2}]},"status":"live","statuses":["live"]}},"name":"Mistral-Small-3.2-24B-Instruct-2506","new":false,"openapi_url":"https://mistral-small-3-2-24b-instruct-2506.endpoints.kepler.ai.cloud.ovh.net/doc/openapi.json","router_openapi_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/doc/Mistral-Small-3.2-24B-Instruct-2506/openapi.json","tags":["Vision"],"version":"v3.1"},{"available":true,"category":"Embeddings","category_v2":"Embeddings","created_at":"2024-11-18T15:34:10Z","description":"The BGE-M3 model, developed by the BAAI and released in early 2024, is a powerful multilingual and multigranular embedding model. It supports texts of varying lengths across over 100 languages and produces dense embeddings optimized for semantic search and retrieval applications.\n","documentation_url":"https://bge-m3.endpoints.kepler.ai.cloud.ovh.net/doc/readme","gradio_url":"https://playground.endpoints.ai.ovh.net/bge-m3","id":"bge-m3","isNew":false,"licence_url":"https://bge-m3.endpoints.kepler.ai.cloud.ovh.net/doc/license","metadata":{"aliases":["BAAI/bge-m3","BGE-M3"],"api_base_url":"https://bge-m3.endpoints.kepler.ai.cloud.ovh.net/api/","benchmark_results":[{"benchmark_name":"mteb_mean_ask","benchmark_score":59.54},{"benchmark_name":"default","benchmark_score":59.54}],"context_size":8192,"deployment_information":{"datacenters":[{"country_code":"FR"}]},"model_base_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/v1","model_specs":{"capabilities":{"batch":true,"fill_in_the_middle":false,"function_calling":false,"input_modality":["text"],"output_modality":["embedding_vector"],"reasoning":false,"response_format":[],"streaming":false,"use_case_tag":["multilingual"]},"model_parameters":{"parameters_type":"embedding","params":{"context_size":8192,"max_batch_size":25,"output_dimension":1024}},"nb_of_billion_parameters":0.567,"quantization":"fp16"},"playground_url":"https://playground.endpoints.ai.ovh.net/bge-m3","publishing_information":{"endpoint_publication_date":"07/11/2024","knowledge_cutoff":"None","languages":["en"],"licence_name":"MIT","licence_url":"https://bge-m3.endpoints.kepler.ai.cloud.ovh.net/doc/license","model_eol_date":"None","model_publication_date":"27/01/2024","publisher":"BAAI","series":"bge","source":"huggingface","source_link":"https://huggingface.co/BAAI/bge-m3"},"router_base_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/v1","tags":[""],"usage_information":{"avg_throughtput":1,"avg_throughtput_unit":"empirical","pricing":[{"price":0.01,"price_unit":"million_input_tokens"}],"rate_limit":{"rate_limit_api":[{"rate_limit_unit":"req/m","rate_limit_value":400}],"rate_limit_api_anonymous":[{"rate_limit_unit":"req/m","rate_limit_value":2}],"rate_limit_playground":[{"rate_limit_unit":"req/m","rate_limit_value":2}]},"status":"live","statuses":["live"]}},"name":"bge-m3","new":false,"openapi_url":"https://bge-m3.endpoints.kepler.ai.cloud.ovh.net/doc/openapi.json","router_openapi_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/doc/bge-m3/openapi.json","tags":[],"version":"v1.0"},{"available":true,"category":"Large Language Models (LLM)","category_v2":"Text Generation","created_at":"2025-04-01T15:10:31Z","description":"Llama 3.3 is an instruction-tuned generative language model optimized for multilingual dialogue use cases. It was released by Meta AI on December 6, 2024, and utilizes an advanced transformer architecture and is designed to align with human preferences for helpfulness and safety through supervised fine-tuning (SFT) and reinforcement learning with human feedback (RLHF). The knowledge cutoff for this 70B model is December 31, 2023.\n","documentation_url":"https://llama-3-3-70b-instruct.endpoints.kepler.ai.cloud.ovh.net/doc/readme","gradio_url":"https://playground.endpoints.ai.ovh.net/Meta-Llama-3_3-70B-Instruct","id":"llama-3-3-70b-instruct","isNew":false,"licence_url":"https://llama-3-3-70b-instruct.endpoints.kepler.ai.cloud.ovh.net/doc/license","metadata":{"aliases":["meta-llama/Llama-3.3-70B-Instruct","meta-llama-3_3-70b-instruct"],"api_base_url":"https://llama-3-3-70b-instruct.endpoints.kepler.ai.cloud.ovh.net/api/openai_compat/","api_model_name":"Meta-Llama-3_3-70B-Instruct","benchmark_results":[{"benchmark_name":"MMLU-Pro","benchmark_score":48},{"benchmark_name":"default","benchmark_score":44.85}],"context":"131k","deployment_information":{"datacenters":[{"country_code":"FR"}]},"model_base_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/v1","model_specs":{"capabilities":{"batch":true,"fill_in_the_middle":false,"function_calling":true,"input_modality":["text"],"output_modality":["text"],"reasoning":false,"response_format":["raw_text","json_object","json_schema"],"streaming":true,"supported_sampling_parameters":["temperature","stop","top_p","seed"],"use_case_tag":["code_chat","code_completion","code_explanation","emotion_detection","entity_extraction","multi_language_coding","multilingual","paraphrasing","sentiment_analysis","summarization","text_generation","text_only_compatible","translation"],"web_search":false},"model_parameters":{"parameters_type":"llm","params":{"context_size_k":131,"max_output_token_k":131}},"nb_of_billion_parameters":70,"quantization":"fp8"},"playground_url":"https://playground.endpoints.ai.ovh.net/Meta-Llama-3_3-70B-Instruct","publishing_information":{"endpoint_publication_date":"20/02/2025","knowledge_cutoff":"31/12/2023","languages":["*"],"licence_name":"Llama 3.3 Community","licence_url":"https://llama-3-3-70b-instruct.endpoints.kepler.ai.cloud.ovh.net/doc/license","model_eol_date":"None","model_publication_date":"06/12/2024","publisher":"meta-llama","series":"llama","source":"huggingface","source_link":"https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct"},"router_base_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/v1","tags":["Streaming"],"usage_information":{"avg_throughtput":2,"avg_throughtput_unit":"empirical","pricing":[{"price":0.67,"price_unit":"million_input_tokens"},{"price":0.67,"price_unit":"million_output_tokens"}],"rate_limit":{"rate_limit_api":[{"rate_limit_unit":"req/m","rate_limit_value":400}],"rate_limit_api_anonymous":[{"rate_limit_unit":"req/m","rate_limit_value":2}],"rate_limit_playground":[{"rate_limit_unit":"req/m","rate_limit_value":2}]},"status":"live","statuses":["live"]}},"name":"Meta-Llama-3_3-70B-Instruct","new":false,"openapi_url":"https://llama-3-3-70b-instruct.endpoints.kepler.ai.cloud.ovh.net/doc/openapi.json","router_openapi_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/doc/Meta-Llama-3_3-70B-Instruct/openapi.json","tags":[],"version":"v1.0"},{"available":false,"category":"Embeddings","category_v2":"Embeddings","created_at":"2025-03-24T14:27:22Z","description":"BGE-Multilingual-Gemma2 is a LLM-based multilingual embedding model. It is trained on a diverse range of languages and tasks. BGE-Multilingual-Gemma2 primarily demonstrates the following advancements:\nDiverse training data: The model's training data spans a broad range of languages, including English, Chinese, Japanese, Korean, French, and more.Additionally, the data covers a variety of task types, such as retrieval, classification, and clustering.\nOutstanding performance: The model exhibits state-of-the-art (SOTA) results on multilingual benchmarks like MIRACL, MTEB-pl, and MTEB-fr. It also achieves excellent performance on other major evaluations, including MTEB, C-MTEB and AIR-Bench.\n","documentation_url":"https://bge-multilingual-gemma2.endpoints.kepler.ai.cloud.ovh.net/doc/readme","gradio_url":"https://playground.endpoints.ai.ovh.net/bge-multilingual-gemma2","id":"bge-multilingual-gemma2","isNew":false,"licence_url":"https://bge-multilingual-gemma2.endpoints.kepler.ai.cloud.ovh.net/doc/license","metadata":{"aliases":["BAAI/bge-multilingual-gemma2"],"api_base_url":"https://bge-multilingual-gemma2.endpoints.kepler.ai.cloud.ovh.net/api/","benchmark_results":[{"benchmark_name":"mteb_mean_ask","benchmark_score":59.54},{"benchmark_name":"default","benchmark_score":59.54}],"context_size":8192,"deployment_information":{"datacenters":[{"country_code":"FR"}]},"model_base_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/v1","model_specs":{"capabilities":{"batch":true,"fill_in_the_middle":false,"function_calling":false,"input_modality":["text"],"output_modality":["embedding_vector"],"reasoning":false,"response_format":[],"streaming":false,"use_case_tag":["multilingual"]},"model_parameters":{"parameters_type":"embedding","params":{"context_size":8192,"max_batch_size":25,"output_dimension":3584}},"nb_of_billion_parameters":0.567,"quantization":"fp16"},"playground_url":"https://playground.endpoints.ai.ovh.net/bge-multilingual-gemma2","publishing_information":{"endpoint_publication_date":"24/03/2025","knowledge_cutoff":"None","languages":["en"],"licence_name":"Gemma","licence_url":"https://bge-multilingual-gemma2.endpoints.kepler.ai.cloud.ovh.net/doc/license","model_eol_date":"21/01/2027","model_publication_date":"29/06/2024","publisher":"BAAI","series":"bge","source":"huggingface","source_link":"https://huggingface.co/BAAI/bge-multilingual-gemma2"},"router_base_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/v1","usage_information":{"avg_throughtput":1,"avg_throughtput_unit":"empirical","pricing":[{"price":0.01,"price_unit":"million_input_tokens"}],"rate_limit":{"rate_limit_api":[{"rate_limit_unit":"req/m","rate_limit_value":400}],"rate_limit_api_anonymous":[{"rate_limit_unit":"req/m","rate_limit_value":2}],"rate_limit_playground":[{"rate_limit_unit":"req/m","rate_limit_value":2}]},"status":"live","statuses":["live"]}},"name":"bge-multilingual-gemma2","new":false,"openapi_url":"https://bge-multilingual-gemma2.endpoints.kepler.ai.cloud.ovh.net/doc/openapi.json","router_openapi_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/doc/bge-multilingual-gemma2/openapi.json","tags":[],"version":"v1.0"},{"available":true,"category":"Reasoning LLM","category_v2":"Text Generation","created_at":"2025-08-28T13:02:25Z","description":"gpt-oss-120b is a cutting-edge model designed for high-level reasoning, instruction-following, and advanced agent capabilities. It excels in versatile use cases, offering powerful performance for both general-purpose and specialized applications.\n","documentation_url":"https://gpt-oss-120b.endpoints.kepler.ai.cloud.ovh.net/doc/readme","gradio_url":"https://playground.endpoints.ai.ovh.net/gpt-oss-120b","id":"gpt-oss-120b","isNew":false,"licence_url":"https://gpt-oss-120b.endpoints.kepler.ai.cloud.ovh.net/doc/license","metadata":{"aliases":["openai/gpt-oss-120b"],"api_base_url":"https://gpt-oss-120b.endpoints.kepler.ai.cloud.ovh.net/api/openai_compat/","benchmark_results":[{"benchmark_name":"MMLU-Pro","benchmark_score":80.8},{"benchmark_name":"LiveCodeBench","benchmark_score":63.9}],"context":"131k","deployment_information":{"datacenters":[{"country_code":"FR"}]},"model_base_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/v1","model_specs":{"capabilities":{"batch":true,"fill_in_the_middle":false,"function_calling":true,"input_modality":["text"],"output_modality":["text"],"reasoning":true,"reasoning_efforts":["low","medium","high"],"response_format":["raw_text","json_object","json_schema"],"streaming":true,"use_case_tag":["code_chat","code_completion","code_explanation","emotion_detection","entity_extraction","multi_language_coding","multilingual","paraphrasing","sentiment_analysis","summarization","text_generation","text_only_compatible","translation"]},"model_parameters":{"parameters_type":"llm","params":{"context_size_k":131,"max_output_token_k":131}},"nb_of_billion_parameters":117,"quantization":"fp4"},"playground_url":"https://playground.endpoints.ai.ovh.net/gpt-oss-120b","publishing_information":{"endpoint_publication_date":"12/08/2025","knowledge_cutoff":"31/06/2024","languages":["*"],"licence_name":"Apache 2.0","licence_url":"https://gpt-oss-120b.endpoints.kepler.ai.cloud.ovh.net/doc/license","model_eol_date":"None","model_publication_date":"05/08/2025","publisher":"OpenAI","series":"gpt-oss","source":"huggingface","source_link":"https://huggingface.co/openai/gpt-oss-120b"},"router_base_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/v1","tags":["Reasoning","Streaming"],"usage_information":{"avg_throughtput":2,"avg_throughtput_unit":"empirical","pricing":[{"price":0.08,"price_unit":"million_input_tokens"},{"price":0.4,"price_unit":"million_output_tokens"}],"rate_limit":{"rate_limit_api":[{"rate_limit_unit":"req/m","rate_limit_value":400}],"rate_limit_api_anonymous":[{"rate_limit_unit":"req/m","rate_limit_value":2}],"rate_limit_playground":[{"rate_limit_unit":"req/m","rate_limit_value":2}]},"status":"live","statuses":["live"]}},"name":"gpt-oss-120b","new":false,"openapi_url":"https://gpt-oss-120b.endpoints.kepler.ai.cloud.ovh.net/doc/openapi.json","router_openapi_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/doc/gpt-oss-120b/openapi.json","tags":["Reasoning"],"version":"v1"},{"available":true,"category":"Text To Speech","category_v2":"Text To Speech","created_at":"2024-10-21T13:04:36Z","description":"This NVIDIA TTS model generates natural-sounding German speech from raw text without requiring additional information.\n","documentation_url":"https://nvr-tts-de-de.endpoints.kepler.ai.cloud.ovh.net/doc/readme","gradio_url":"https://playground.endpoints.ai.ovh.net/nvr-tts-de-de","id":"nvr-tts-de-de","isNew":false,"licence_url":"https://nvr-tts-de-de.endpoints.kepler.ai.cloud.ovh.net/doc/license","metadata":{"aliases":["NVIDIA-RIVA"],"api_base_url":"https://nvr-tts-de-de.endpoints.kepler.ai.cloud.ovh.net/api/v1/","deployment_information":{"datacenters":[{"country_code":"FR"}]},"model_base_url":"https://nvr-tts-de-de.endpoints.kepler.ai.cloud.ovh.net/api/v1","model_specs":{"capabilities":{"batch":false,"fill_in_the_middle":false,"function_calling":false,"input_modality":["text"],"output_modality":["sound"],"reasoning":false,"response_format":["audio"],"streaming":false,"use_case_tag":["text-to-speech"]},"model_parameters":{"parameters_type":"tts","params":{"input_languages":["de-de"],"output_languages":["de-de"]}},"quantization":"fp32"},"playground_url":"https://playground.endpoints.ai.ovh.net/nvr-tts-de-de","publishing_information":{"endpoint_publication_date":"28/06/2024","knowledge_cutoff":"None","languages":["de-de"],"licence_name":"Riva license","licence_url":"https://nvr-tts-de-de.endpoints.kepler.ai.cloud.ovh.net/doc/license","model_eol_date":"None","model_publication_date":"None","publisher":"NVIDIA Riva","series":"NVIDIA Riva","source":"NVIDIA"},"router_base_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/v1","usage_information":{"avg_throughtput":1,"avg_throughtput_unit":"empirical","pricing":[{"price":0,"price_unit":"million_input_chars"}],"rate_limit":{"rate_limit_api":[{"rate_limit_unit":"req/m","rate_limit_value":400}],"rate_limit_api_anonymous":[{"rate_limit_unit":"req/m","rate_limit_value":2}],"rate_limit_playground":[{"rate_limit_unit":"req/m","rate_limit_value":2}]},"status":"live","statuses":["live"]}},"name":"nvr-tts-de-de","new":false,"openapi_url":"https://nvr-tts-de-de.endpoints.kepler.ai.cloud.ovh.net/doc/openapi.json","router_openapi_url":"https://nvr-tts-de-de.endpoints.kepler.ai.cloud.ovh.net/doc/openapi.json","tags":["Text To Speech","Multi-speaker"],"version":"v1.0"},{"available":true,"category":"Visual LLM","category_v2":"Text Generation","created_at":"2025-03-31T12:26:23Z","description":"Qwen2.5-VL is a powerful vision-language model, designed for advanced image understanding. It can generate detailed image captions, analyze documents, OCR, detect objects, and answer questions based on visuals, making it useful for AI assistants, RAG and Agents.","documentation_url":"https://qwen-2-5-vl-72b-instruct.endpoints.kepler.ai.cloud.ovh.net/doc/readme","gradio_url":"https://playground.endpoints.ai.ovh.net/Qwen2.5-VL-72B-Instruct","id":"qwen-2-5-vl-72b-instruct","isNew":false,"licence_url":"https://qwen-2-5-vl-72b-instruct.endpoints.kepler.ai.cloud.ovh.net/doc/license","metadata":{"aliases":["Qwen/Qwen2.5-VL-72B-Instruct","qwen2.5-vl-72b-instruct"],"api_base_url":"https://qwen-2-5-vl-72b-instruct.endpoints.kepler.ai.cloud.ovh.net/api/openai_compat/","benchmark_results":[{"benchmark_name":"AVG OpenVLM","benchmark_score":76.1},{"benchmark_name":"default","benchmark_score":76.1}],"context":"32k","deployment_information":{"datacenters":[{"country_code":"FR"}]},"model_base_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/v1","model_specs":{"capabilities":{"batch":true,"fill_in_the_middle":false,"function_calling":false,"input_modality":["text","image"],"output_modality":["text"],"reasoning":false,"response_format":["raw_text","json_object","json_schema"],"streaming":true,"supported_sampling_parameters":["temperature","stop","top_p","seed"],"use_case_tag":["multimodal","vision","text_only_compatible"],"web_search":false},"model_parameters":{"parameters_type":"vlm","params":{"context_size_k":32,"max_images_per_prompt":5,"max_output_token_k":32,"supported_aspect_ratios":["*"]}},"nb_of_billion_parameters":72,"quantization":"fp8"},"playground_url":"https://playground.endpoints.ai.ovh.net/Qwen2.5-VL-72B-Instruct","publishing_information":{"endpoint_publication_date":"26/03/2025","knowledge_cutoff":"None","languages":["*"],"licence_name":"Qwen","licence_url":"https://qwen-2-5-vl-72b-instruct.endpoints.kepler.ai.cloud.ovh.net/doc/license","model_eol_date":"None","model_publication_date":"27/01/2025","publisher":"Qwen","series":"qwen","source":"huggingface","source_link":"https://huggingface.co/Qwen/Qwen2.5-VL-72B-Instruct"},"router_base_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/v1","tags":["Vision","Streaming"],"usage_information":{"avg_throughtput":2,"avg_throughtput_unit":"empirical","pricing":[{"price":0.91,"price_unit":"million_input_tokens"},{"price":0.91,"price_unit":"million_output_tokens"}],"rate_limit":{"rate_limit_api":[{"rate_limit_unit":"req/m","rate_limit_value":400}],"rate_limit_api_anonymous":[{"rate_limit_unit":"req/m","rate_limit_value":2}],"rate_limit_playground":[{"rate_limit_unit":"req/m","rate_limit_value":2}]},"status":"live","statuses":["live"]}},"name":"Qwen2.5-VL-72B-Instruct","new":false,"openapi_url":"https://qwen-2-5-vl-72b-instruct.endpoints.kepler.ai.cloud.ovh.net/doc/openapi.json","router_openapi_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/doc/Qwen2.5-VL-72B-Instruct/openapi.json","tags":["Vision"],"version":"v2.5"},{"available":true,"category":"Visual LLM","category_v2":"Text Generation","created_at":"2026-06-01T08:32:14Z","description":"Qwen3.6-27B is a 27-billion-parameter multimodal causal language model developed by Alibaba's Qwen team, released as an open-weight variant following the February release of the Qwen3.5 series. It is built as an Image-Text-to-Text model.\n","documentation_url":"https://qwen-3-6-27b.endpoints.kepler.ai.cloud.ovh.net/doc/readme","gradio_url":"https://playground.endpoints.ai.ovh.net/Qwen3.6-27B","id":"qwen-3-6-27b","isNew":true,"licence_url":"https://qwen-3-6-27b.endpoints.kepler.ai.cloud.ovh.net/doc/license","metadata":{"aliases":["Qwen/Qwen3.6-27B","qwen3.6-27b"],"api_base_url":"https://qwen-3-6-27b.endpoints.kepler.ai.cloud.ovh.net/api/openai_compat/","benchmark_results":[{"benchmark_name":"MMLU-Pro","benchmark_score":85.2}],"context":"262k","deployment_information":{"datacenters":[{"country_code":"FR"}]},"model_base_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/v1","model_specs":{"capabilities":{"batch":true,"fill_in_the_middle":false,"function_calling":true,"input_modality":["text","image"],"output_modality":["text"],"reasoning":true,"response_format":["raw_text","json_object","json_schema"],"streaming":true,"supported_sampling_parameters":["temperature","stop","top_p","seed"],"use_case_tag":["vision","code_chat","code_completion","code_explanation","emotion_detection","entity_extraction","multi_language_coding","multilingual","paraphrasing","sentiment_analysis","summarization","text_generation","text_only_compatible","translation","multimodal"],"web_search":false},"model_parameters":{"parameters_type":"vlm","params":{"context_size_k":262,"max_images_per_prompt":5,"max_output_token_k":262,"supported_aspect_ratios":["*"]}},"nb_of_billion_parameters":27,"quantization":"fp8"},"playground_url":"https://playground.endpoints.ai.ovh.net/Qwen3.6-27B","publishing_information":{"endpoint_publication_date":"06/05/2026","knowledge_cutoff":"None","languages":["*"],"licence_name":"Apache 2.0","licence_url":"https://qwen-3-6-27b.endpoints.kepler.ai.cloud.ovh.net/doc/license","model_eol_date":"None","model_publication_date":"15/04/2026","publisher":"Qwen","series":"qwen","source":"huggingface","source_link":"https://huggingface.co/Qwen/Qwen3.6-27B"},"router_base_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/v1","tags":["Vision","Reasoning","Streaming"],"usage_information":{"avg_throughtput":2,"avg_throughtput_unit":"empirical","pricing":[{"price":0.4,"price_unit":"million_input_tokens"},{"price":2.7,"price_unit":"million_output_tokens"}],"rate_limit":{"rate_limit_api":[{"rate_limit_unit":"req/m","rate_limit_value":400}],"rate_limit_api_anonymous":[{"rate_limit_unit":"req/m","rate_limit_value":2}],"rate_limit_playground":[{"rate_limit_unit":"req/m","rate_limit_value":2}]},"statuses":["live"]}},"name":"Qwen3.6-27B","new":true,"openapi_url":"https://qwen-3-6-27b.endpoints.kepler.ai.cloud.ovh.net/doc/openapi.json","router_openapi_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/doc/Qwen3.6-27B/openapi.json","tags":["Vision","Reasoning"],"version":"v3.6"},{"available":true,"category":"Visual LLM","category_v2":"Text Generation","created_at":"2026-05-18T14:11:22Z","description":"Qwen3.5-397B-A17B is a highly efficient, reasoning focused MoE (Mixture of Experts) model delivering state-of-the-art performance in language, vision, and multimodal tasks, with strong capabilities in code, reasoning, and real-world adaptability.\n","documentation_url":"https://qwen-3-5-397b.endpoints.kepler.ai.cloud.ovh.net/doc/readme","gradio_url":"https://playground.endpoints.ai.ovh.net/Qwen3.5-397B-A17B","id":"qwen-3-5-397b","isNew":true,"licence_url":"https://qwen-3-5-397b.endpoints.kepler.ai.cloud.ovh.net/doc/license","metadata":{"aliases":["Qwen/Qwen3.5-397B-A17B","qwen3.5-397b-a17b","qwen3.5-397b"],"api_base_url":"https://qwen-3-5-397b.endpoints.kepler.ai.cloud.ovh.net/api/openai_compat/","benchmark_results":[{"benchmark_name":"MMLU-Pro","benchmark_score":87.8},{"benchmark_name":"IFEval","benchmark_score":92.6},{"benchmark_name":"LiveCodeBench-v6","benchmark_score":83.6},{"benchmark_name":"BFCL-V4","benchmark_score":72.9},{"benchmark_name":"SWE-bench-Verified","benchmark_score":76.4},{"benchmark_name":"MMMU-Pro","benchmark_score":79},{"benchmark_name":"OmniDocBench1.5","benchmark_score":90.8},{"benchmark_name":"OCRBench","benchmark_score":93.1}],"context":"262k","deployment_information":{"datacenters":[{"country_code":"FR"}]},"model_base_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/v1","model_specs":{"capabilities":{"batch":true,"fill_in_the_middle":false,"function_calling":true,"input_modality":["text","image"],"output_modality":["text"],"reasoning":true,"response_format":["raw_text","json_object","json_schema"],"streaming":true,"supported_sampling_parameters":["temperature","stop","top_p","seed"],"use_case_tag":["emotion_detection","entity_extraction","multi_language_coding","vision","multilingual","paraphrasing","sentiment_analysis","summarization","text_generation","text_only_compatible","translation","multimodal"],"web_search":false},"model_parameters":{"parameters_type":"vlm","params":{"context_size_k":262,"max_images_per_prompt":5,"max_output_token_k":262,"supported_aspect_ratios":["*"]}},"nb_of_billion_parameters":397,"quantization":"fp8"},"playground_url":"https://playground.endpoints.ai.ovh.net/Qwen3.5-397B-A17B","publishing_information":{"endpoint_publication_date":"15/05/2026","knowledge_cutoff":"31/09/2025","languages":["*"],"licence_name":"Apache 2.0","licence_url":"https://qwen-3-5-397b.endpoints.kepler.ai.cloud.ovh.net/doc/license","model_eol_date":"None","model_publication_date":"15/02/2026","publisher":"Qwen","series":"qwen","source":"huggingface","source_link":"https://huggingface.co/Qwen/Qwen3.5-397B-A17B"},"router_base_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/v1","tags":["Vision","Reasoning","Streaming"],"usage_information":{"avg_throughtput":2,"avg_throughtput_unit":"empirical","pricing":[{"price":0.6,"price_unit":"million_input_tokens"},{"price":3.6,"price_unit":"million_output_tokens"}],"rate_limit":{"rate_limit_api":[{"rate_limit_unit":"req/m","rate_limit_value":400}],"rate_limit_api_anonymous":[{"rate_limit_unit":"req/m","rate_limit_value":2}],"rate_limit_playground":[{"rate_limit_unit":"req/m","rate_limit_value":2}]},"statuses":["live"]}},"name":"Qwen3.5-397B-A17B","new":true,"openapi_url":"https://qwen-3-5-397b.endpoints.kepler.ai.cloud.ovh.net/doc/openapi.json","router_openapi_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/doc/Qwen3.5-397B-A17B/openapi.json","tags":["Vision","Reasoning"],"version":"v3.5"},{"available":true,"category":"Visual LLM","category_v2":"Text Generation","created_at":"2026-04-22T10:24:08Z","description":"Qwen3.5 9B is a highly efficient, reasoning focused model delivering state of the art performance in language, vision, and multimodal tasks, with strong capabilities in code, reasoning, and real-world adaptability.\n","documentation_url":"https://qwen-3-5-9b.endpoints.kepler.ai.cloud.ovh.net/doc/readme","gradio_url":"https://playground.endpoints.ai.ovh.net/Qwen3.5-9B","id":"qwen-3-5-9b","isNew":true,"licence_url":"https://qwen-3-5-9b.endpoints.kepler.ai.cloud.ovh.net/doc/license","metadata":{"aliases":["Qwen/Qwen3.5-9B","qwen3.5-9b"],"api_base_url":"https://qwen-3-5-9b.endpoints.kepler.ai.cloud.ovh.net/api/openai_compat/","benchmark_results":[{"benchmark_name":"MMLU-Pro","benchmark_score":69}],"context":"262k","deployment_information":{"datacenters":[{"country_code":"FR"}]},"model_base_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/v1","model_specs":{"capabilities":{"batch":true,"fill_in_the_middle":false,"function_calling":true,"input_modality":["text","image"],"output_modality":["text"],"reasoning":true,"response_format":["raw_text","json_object","json_schema"],"streaming":true,"supported_sampling_parameters":["temperature","stop","top_p","seed"],"use_case_tag":["vision","code_chat","code_completion","code_explanation","emotion_detection","entity_extraction","multi_language_coding","multilingual","paraphrasing","sentiment_analysis","summarization","text_generation","text_only_compatible","translation","multimodal"],"web_search":false},"model_parameters":{"parameters_type":"vlm","params":{"context_size_k":262,"max_images_per_prompt":5,"max_output_token_k":262,"supported_aspect_ratios":["*"]}},"nb_of_billion_parameters":9.7,"quantization":"bf16"},"playground_url":"https://playground.endpoints.ai.ovh.net/Qwen3.5-9B","publishing_information":{"endpoint_publication_date":"15/04/2026","knowledge_cutoff":"31/09/2025","languages":["*"],"licence_name":"Apache 2.0","licence_url":"https://qwen-3-5-9b.endpoints.kepler.ai.cloud.ovh.net/doc/license","model_eol_date":"None","model_publication_date":"15/02/2026","publisher":"Qwen","series":"qwen","source":"huggingface","source_link":"https://huggingface.co/Qwen/Qwen3.5-9B"},"router_base_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/v1","tags":["Vision","Reasoning","Streaming"],"usage_information":{"avg_throughtput":2,"avg_throughtput_unit":"empirical","pricing":[{"price":0.1,"price_unit":"million_input_tokens"},{"price":0.15,"price_unit":"million_output_tokens"}],"rate_limit":{"rate_limit_api":[{"rate_limit_unit":"req/m","rate_limit_value":400}],"rate_limit_api_anonymous":[{"rate_limit_unit":"req/m","rate_limit_value":2}],"rate_limit_playground":[{"rate_limit_unit":"req/m","rate_limit_value":2}]},"statuses":["live"]}},"name":"Qwen3.5-9B","new":true,"openapi_url":"https://qwen-3-5-9b.endpoints.kepler.ai.cloud.ovh.net/doc/openapi.json","router_openapi_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/doc/Qwen3.5-9B/openapi.json","tags":["Vision","Reasoning"],"version":"v3.5"},{"available":true,"category":"LLM Guard","category_v2":"Safety & Moderation","created_at":"2026-01-22T10:46:12Z","description":"Qwen3Guard is a series of safety moderation models built upon Qwen3 and trained on a dataset of 1.19 million prompts and responses labeled for safety. The series includes models of three sizes (0.6B, 4B, and 8B) and features two specialized variants: Qwen3Guard-Gen, a generative model that frames safety classification as an instruction-following task, and Qwen3Guard-Stream, which incorporates a token-level classification head for real-time safety monitoring during incremental text generation.\n","documentation_url":"https://qwen-guard-gen-06b.endpoints.kepler.ai.cloud.ovh.net/doc/readme","gradio_url":"https://playground.endpoints.ai.ovh.net/Qwen3Guard-Gen-0.6B","id":"qwen-guard-gen-06b","isNew":false,"licence_url":"https://qwen-guard-gen-06b.endpoints.kepler.ai.cloud.ovh.net/doc/license","metadata":{"aliases":["Qwen/Qwen3Guard-Gen-0.6B"],"api_base_url":"https://qwen-guard-gen-06b.endpoints.kepler.ai.cloud.ovh.net/api/openai_compat/","benchmark_results":[{"benchmark_name":"WildGuardTest","benchmark_score":89.4}],"context":"32k","deployment_information":{"datacenters":[{"country_code":"FR"}]},"model_base_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/v1","model_specs":{"capabilities":{"batch":true,"fill_in_the_middle":false,"function_calling":false,"input_modality":["text"],"output_modality":["text"],"reasoning":false,"response_format":["raw_text"],"streaming":true,"supported_sampling_parameters":["temperature","stop","top_p","seed"],"use_case_tag":["moderation"],"web_search":false},"model_parameters":{"parameters_type":"llm","params":{"context_size_k":32,"max_output_token_k":16}},"nb_of_billion_parameters":0.6,"quantization":"fp16"},"playground_url":"https://playground.endpoints.ai.ovh.net/Qwen3Guard-Gen-0.6B","publishing_information":{"endpoint_publication_date":"21/01/2026","knowledge_cutoff":"None","languages":["*"],"licence_name":"Apache 2.0","licence_url":"https://qwen-guard-gen-06b.endpoints.kepler.ai.cloud.ovh.net/doc/license","model_eol_date":"None","model_publication_date":"23/09/2025","publisher":"qwen3guard","series":"qwen3guard","source":"huggingface","source_link":"https://huggingface.co/Qwen/Qwen3Guard-Gen-0.6B"},"router_base_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/v1","tags":["Moderation","Streaming"],"usage_information":{"avg_throughtput":1,"avg_throughtput_unit":"empirical","pricing":[{"price":0,"price_unit":"million_input_tokens"},{"price":0,"price_unit":"million_output_tokens"}],"rate_limit":{"rate_limit_api":[{"rate_limit_unit":"req/m","rate_limit_value":400}],"rate_limit_api_anonymous":[{"rate_limit_unit":"req/m","rate_limit_value":2}],"rate_limit_playground":[{"rate_limit_unit":"req/m","rate_limit_value":2}]},"status":"live","statuses":["beta"]}},"name":"Qwen3Guard-Gen-0.6B","new":false,"openapi_url":"https://qwen-guard-gen-06b.endpoints.kepler.ai.cloud.ovh.net/doc/openapi.json","router_openapi_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/doc/Qwen3Guard-Gen-0.6B/openapi.json","tags":["Moderation"],"version":"v3.1"},{"available":true,"category":"Text To Speech","category_v2":"Text To Speech","created_at":"2024-10-21T13:04:36Z","description":"This NVIDIA TTS model generates natural-sounding European Spanish speech from raw text without requiring additional information.\n","documentation_url":"https://nvr-tts-es-es.endpoints.kepler.ai.cloud.ovh.net/doc/readme","gradio_url":"https://playground.endpoints.ai.ovh.net/nvr-tts-es-es","id":"nvr-tts-es-es","isNew":false,"licence_url":"https://nvr-tts-es-es.endpoints.kepler.ai.cloud.ovh.net/doc/license","metadata":{"aliases":["NVIDIA-RIVA"],"api_base_url":"https://nvr-tts-es-es.endpoints.kepler.ai.cloud.ovh.net/api/v1/","deployment_information":{"datacenters":[{"country_code":"FR"}]},"model_base_url":"https://nvr-tts-es-es.endpoints.kepler.ai.cloud.ovh.net/api/v1","model_specs":{"capabilities":{"batch":false,"fill_in_the_middle":false,"function_calling":false,"input_modality":["text"],"output_modality":["sound"],"reasoning":false,"response_format":["audio"],"streaming":false,"use_case_tag":["text-to-speech","multi-speaker"]},"model_parameters":{"parameters_type":"tts","params":{"input_languages":["es-es"],"output_languages":["es-es"]}},"quantization":"fp32"},"playground_url":"https://playground.endpoints.ai.ovh.net/nvr-tts-es-es","publishing_information":{"endpoint_publication_date":"28/06/2024","knowledge_cutoff":"None","languages":["es-es"],"licence_name":"Riva license","licence_url":"https://nvr-tts-es-es.endpoints.kepler.ai.cloud.ovh.net/doc/license","model_eol_date":"None","model_publication_date":"None","publisher":"NVIDIA Riva","series":"NVIDIA Riva","source":"NVIDIA"},"router_base_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/v1","usage_information":{"avg_throughtput":1,"avg_throughtput_unit":"empirical","pricing":[{"price":0,"price_unit":"million_input_chars"}],"rate_limit":{"rate_limit_api":[{"rate_limit_unit":"req/m","rate_limit_value":400}],"rate_limit_api_anonymous":[{"rate_limit_unit":"req/m","rate_limit_value":2}],"rate_limit_playground":[{"rate_limit_unit":"req/m","rate_limit_value":2}]},"status":"live","statuses":["live"]}},"name":"nvr-tts-es-es","new":false,"openapi_url":"https://nvr-tts-es-es.endpoints.kepler.ai.cloud.ovh.net/doc/openapi.json","router_openapi_url":"https://nvr-tts-es-es.endpoints.kepler.ai.cloud.ovh.net/doc/openapi.json","tags":["Text To Speech","Multi-speaker"],"version":"v1.0"},{"available":true,"category":"Visual LLM","category_v2":"Text Generation","created_at":"2026-09-02T13:41:25Z","description":"Qwen3.8-27B is a 27-billion-parameter multimodal causal language model developed by Alibaba's Qwen team. It is the most capable generation in the Qwen open-model family, built as a native vision-language model that understands images and videos, with flexible thinking control, designed to carry complex, multi-step agentic tasks through to completion with greater reliability.\n","documentation_url":"https://qwen-3-8-27b.endpoints.kepler.ai.cloud.ovh.net/doc/readme","gradio_url":"https://playground.endpoints.ai.ovh.net/Qwen3.8-27B","id":"qwen-3-8-27b","isNew":true,"licence_url":"https://qwen-3-8-27b.endpoints.kepler.ai.cloud.ovh.net/doc/license","metadata":{"aliases":["Qwen/Qwen3.8-27B","qwen3.8-27b"],"api_base_url":"https://qwen-3-8-27b.endpoints.kepler.ai.cloud.ovh.net/api/openai_compat/","benchmark_results":[{"benchmark_name":"SWE-bench Pro","benchmark_score":61.7},{"benchmark_name":"LiveCodeBench v6","benchmark_score":90.3},{"benchmark_name":"GPQA Diamond","benchmark_score":89.2},{"benchmark_name":"OSWorld-Verified","benchmark_score":84.3}],"context":"262k","deployment_information":{"datacenters":[{"country_code":"FR"}]},"model_base_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/v1","model_specs":{"capabilities":{"batch":true,"fill_in_the_middle":false,"function_calling":true,"input_modality":["text","image"],"output_modality":["text"],"reasoning":true,"reasoning_efforts":["none","low","medium","xhigh"],"response_format":["raw_text","json_object","json_schema"],"streaming":true,"supported_sampling_parameters":["temperature","stop","top_p","seed"],"use_case_tag":["vision","code_chat","code_completion","code_explanation","emotion_detection","entity_extraction","multi_language_coding","multilingual","paraphrasing","sentiment_analysis","summarization","text_generation","text_only_compatible","translation","multimodal"],"web_search":false},"model_parameters":{"parameters_type":"vlm","params":{"context_size_k":262,"max_images_per_prompt":5,"max_output_token_k":262,"supported_aspect_ratios":["*"]}},"nb_of_billion_parameters":27,"quantization":"fp8"},"playground_url":"https://playground.endpoints.ai.ovh.net/Qwen3.8-27B","publishing_information":{"endpoint_publication_date":"24/08/2026","knowledge_cutoff":"None","languages":["*"],"licence_name":"Apache 2.0","licence_url":"https://qwen-3-8-27b.endpoints.kepler.ai.cloud.ovh.net/doc/license","model_eol_date":"None","model_publication_date":"01/08/2026","publisher":"Qwen","series":"qwen","source":"huggingface","source_link":"https://huggingface.co/Qwen/Qwen3.8-27B"},"router_base_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/v1","tags":["Vision","Reasoning","Streaming"],"usage_information":{"avg_throughtput":2,"avg_throughtput_unit":"empirical","pricing":[{"price":0.4,"price_unit":"million_input_tokens"},{"price":2.7,"price_unit":"million_output_tokens"}],"rate_limit":{"rate_limit_api":[{"rate_limit_unit":"req/m","rate_limit_value":400}],"rate_limit_api_anonymous":[{"rate_limit_unit":"req/m","rate_limit_value":2}],"rate_limit_playground":[{"rate_limit_unit":"req/m","rate_limit_value":2}]},"statuses":["live"]}},"name":"Qwen3.8-27B","new":true,"openapi_url":"https://qwen-3-8-27b.endpoints.kepler.ai.cloud.ovh.net/doc/openapi.json","router_openapi_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/doc/Qwen3.8-27B/openapi.json","tags":["Vision","Reasoning"],"version":"v3.8"},{"available":true,"category":"Embeddings","category_v2":"Embeddings","created_at":"2026-03-26T13:55:26Z","description":"State-of-the-art text embedding model with 100+ language support and customizable dimensional outputs. Perfect for retrieval, classification, and multilingual applications. The flexible output dimensionality lets you optimize for your specific use case, whether you need compact vectors for fast retrieval or high dimensional representations for complex semantic tasks.\n","documentation_url":"https://qwen3-embedding-8b.endpoints.kepler.ai.cloud.ovh.net/doc/readme","gradio_url":"https://playground.endpoints.ai.ovh.net/Qwen3-Embedding-8B","id":"qwen3-embedding-8b","isNew":true,"licence_url":"https://qwen3-embedding-8b.endpoints.kepler.ai.cloud.ovh.net/doc/license","metadata":{"aliases":["Qwen/qwen3_embedding_8b"],"api_base_url":"https://qwen3-embedding-8b.endpoints.kepler.ai.cloud.ovh.net/api/","benchmark_results":[{"benchmark_name":"mteb_multilingual_mean_task","benchmark_score":70.58},{"benchmark_name":"mteb_multilingual_mean_task_type","benchmark_score":61.69},{"benchmark_name":"mteb_multilingual_bitext_mining","benchmark_score":80.89},{"benchmark_name":"mteb_multilingual_classification","benchmark_score":74},{"benchmark_name":"mteb_multilingual_clustering","benchmark_score":57.65},{"benchmark_name":"mteb_multilingual_instruction_reranking","benchmark_score":10.06},{"benchmark_name":"mteb_multilingual_multilabel_classification","benchmark_score":28.66},{"benchmark_name":"mteb_multilingual_pair_classification","benchmark_score":86.4},{"benchmark_name":"mteb_multilingual_reranking","benchmark_score":65.63},{"benchmark_name":"mteb_multilingual_retrieval","benchmark_score":70.88},{"benchmark_name":"mteb_multilingual_sts","benchmark_score":81.08}],"context":"32k","deployment_information":{"datacenters":[{"country_code":"FR"}]},"model_base_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/v1","model_specs":{"capabilities":{"batch":true,"fill_in_the_middle":false,"function_calling":false,"input_modality":["text"],"output_modality":["embedding_vector"],"reasoning":false,"response_format":[],"streaming":false,"supported_sampling_parameters":[],"use_case_tag":["multilingual"],"web_search":false},"model_parameters":{"parameters_type":"embedding","params":{"context_size":32768,"max_batch_size":25,"output_dimension":4096,"output_dimension_max":4096,"output_dimension_min":32}},"nb_of_billion_parameters":7.6,"quantization":"fp16"},"playground_url":"https://playground.endpoints.ai.ovh.net/Qwen3-Embedding-8B","publishing_information":{"endpoint_publication_date":"26/03/2026","knowledge_cutoff":"None","languages":["*"],"licence_name":"Apache 2.0","licence_url":"https://qwen3-embedding-8b.endpoints.kepler.ai.cloud.ovh.net/doc/license","model_eol_date":"None","model_publication_date":"06/06/2025","publisher":"Qwen","series":"qwen","source":"huggingface","source_link":"https://huggingface.co/Qwen/qwen3_embedding_8b"},"router_base_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/v1","tags":[""],"usage_information":{"avg_throughtput":1,"avg_throughtput_unit":"empirical","pricing":[{"price":0.1,"price_unit":"million_input_tokens"}],"rate_limit":{"rate_limit_api":[{"rate_limit_unit":"req/m","rate_limit_value":400}],"rate_limit_api_anonymous":[{"rate_limit_unit":"req/m","rate_limit_value":2}],"rate_limit_playground":[{"rate_limit_unit":"req/m","rate_limit_value":2}]},"statuses":["live"]}},"name":"Qwen3-Embedding-8B","new":true,"openapi_url":"https://qwen3-embedding-8b.endpoints.kepler.ai.cloud.ovh.net/doc/openapi.json","router_openapi_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/doc/Qwen3-Embedding-8B/openapi.json","tags":[],"version":"v1.0"},{"available":true,"category":"LLM Guard","category_v2":"Safety & Moderation","created_at":"2026-01-22T10:46:13Z","description":"Qwen3Guard is a series of safety moderation models built upon Qwen3 and trained on a dataset of 1.19 million prompts and responses labeled for safety. The series includes models of three sizes (0.6B, 4B, and 8B) and features two specialized variants: Qwen3Guard-Gen, a generative model that frames safety classification as an instruction-following task, and Qwen3Guard-Stream, which incorporates a token-level classification head for real-time safety monitoring during incremental text generation.\n","documentation_url":"https://qwen-guard-gen-8b.endpoints.kepler.ai.cloud.ovh.net/doc/readme","gradio_url":"https://playground.endpoints.ai.ovh.net/Qwen3Guard-Gen-8B","id":"qwen-guard-gen-8b","isNew":false,"licence_url":"https://qwen-guard-gen-8b.endpoints.kepler.ai.cloud.ovh.net/doc/license","metadata":{"aliases":["Qwen/Qwen3Guard-Gen-8B"],"api_base_url":"https://qwen-guard-gen-8b.endpoints.kepler.ai.cloud.ovh.net/api/openai_compat/","benchmark_results":[{"benchmark_name":"WildGuardTest","benchmark_score":89.4}],"context":"32k","deployment_information":{"datacenters":[{"country_code":"FR"}]},"model_base_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/v1","model_specs":{"capabilities":{"batch":true,"fill_in_the_middle":false,"function_calling":false,"input_modality":["text"],"output_modality":["text"],"reasoning":false,"response_format":["raw_text"],"streaming":true,"supported_sampling_parameters":["temperature","stop","top_p","seed"],"use_case_tag":["moderation"],"web_search":false},"model_parameters":{"parameters_type":"llm","params":{"context_size_k":32,"max_output_token_k":16}},"nb_of_billion_parameters":8,"quantization":"fp16"},"playground_url":"https://playground.endpoints.ai.ovh.net/Qwen3Guard-Gen-8B","publishing_information":{"endpoint_publication_date":"21/01/2026","knowledge_cutoff":"None","languages":["*"],"licence_name":"Apache 2.0","licence_url":"https://qwen-guard-gen-8b.endpoints.kepler.ai.cloud.ovh.net/doc/license","model_eol_date":"None","model_publication_date":"23/09/2025","publisher":"qwen3guard","series":"qwen3guard","source":"huggingface","source_link":"https://huggingface.co/Qwen/Qwen3Guard-Gen-8B"},"router_base_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/v1","tags":["Moderation","Streaming"],"usage_information":{"avg_throughtput":1,"avg_throughtput_unit":"empirical","pricing":[{"price":0,"price_unit":"million_input_tokens"},{"price":0,"price_unit":"million_output_tokens"}],"rate_limit":{"rate_limit_api":[{"rate_limit_unit":"req/m","rate_limit_value":400}],"rate_limit_api_anonymous":[{"rate_limit_unit":"req/m","rate_limit_value":2}],"rate_limit_playground":[{"rate_limit_unit":"req/m","rate_limit_value":2}]},"status":"live","statuses":["beta"]}},"name":"Qwen3Guard-Gen-8B","new":false,"openapi_url":"https://qwen-guard-gen-8b.endpoints.kepler.ai.cloud.ovh.net/doc/openapi.json","router_openapi_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/doc/Qwen3Guard-Gen-8B/openapi.json","tags":["Moderation"],"version":"v3.1"},{"available":true,"category":"Text To Speech","category_v2":"Text To Speech","created_at":"2024-10-21T13:04:36Z","description":"This NVIDIA TTS model generates natural-sounding Italian speech from raw text without requiring additional information.\n","documentation_url":"https://nvr-tts-it-it.endpoints.kepler.ai.cloud.ovh.net/doc/readme","gradio_url":"https://playground.endpoints.ai.ovh.net/nvr-tts-it-it","id":"nvr-tts-it-it","isNew":false,"licence_url":"https://nvr-tts-it-it.endpoints.kepler.ai.cloud.ovh.net/doc/license","metadata":{"aliases":["NVIDIA-RIVA"],"api_base_url":"https://nvr-tts-it-it.endpoints.kepler.ai.cloud.ovh.net/api/v1/","deployment_information":{"datacenters":[{"country_code":"FR"}]},"model_base_url":"https://nvr-tts-it-it.endpoints.kepler.ai.cloud.ovh.net/api/v1","model_specs":{"capabilities":{"batch":false,"fill_in_the_middle":false,"function_calling":false,"input_modality":["text"],"output_modality":["sound"],"reasoning":false,"response_format":["audio"],"streaming":false,"use_case_tag":["text-to-speech","multi-speaker"]},"model_parameters":{"parameters_type":"tts","params":{"input_languages":["it-it"],"output_languages":["it-it"]}},"quantization":"fp32"},"playground_url":"https://playground.endpoints.ai.ovh.net/nvr-tts-it-it","publishing_information":{"endpoint_publication_date":"28/06/2024","knowledge_cutoff":"None","languages":["it-it"],"licence_name":"Riva license","licence_url":"https://nvr-tts-it-it.endpoints.kepler.ai.cloud.ovh.net/doc/license","model_eol_date":"None","model_publication_date":"None","publisher":"NVIDIA Riva","series":"NVIDIA Riva","source":"NVIDIA"},"router_base_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/v1","usage_information":{"avg_throughtput":1,"avg_throughtput_unit":"empirical","pricing":[{"price":0,"price_unit":"million_input_chars"}],"rate_limit":{"rate_limit_api":[{"rate_limit_unit":"req/m","rate_limit_value":400}],"rate_limit_api_anonymous":[{"rate_limit_unit":"req/m","rate_limit_value":2}],"rate_limit_playground":[{"rate_limit_unit":"req/m","rate_limit_value":2}]},"status":"live","statuses":["live"]}},"name":"nvr-tts-it-it","new":false,"openapi_url":"https://nvr-tts-it-it.endpoints.kepler.ai.cloud.ovh.net/doc/openapi.json","router_openapi_url":"https://nvr-tts-it-it.endpoints.kepler.ai.cloud.ovh.net/doc/openapi.json","tags":["Text To Speech","Multi-speaker"],"version":"v1.0"},{"available":true,"category":"Image Generation","category_v2":"Image Generation","created_at":"2024-10-30T14:13:18Z","description":"SDXL, developed by Stability AI, is an advanced text-to-image model released in July 2023, offering enhanced image generation capabilities.\n","documentation_url":"https://stable-diffusion-xl.endpoints.kepler.ai.cloud.ovh.net/doc/readme","gradio_url":"https://playground.endpoints.ai.ovh.net/stable-diffusion-xl-base-v10","id":"stable-diffusion-xl","isNew":false,"licence_url":"https://stable-diffusion-xl.endpoints.kepler.ai.cloud.ovh.net/doc/license","metadata":{"aliases":["stabilityai/stable-diffusion-xl-base-1.0"],"api_base_url":"https://stable-diffusion-xl.endpoints.kepler.ai.cloud.ovh.net/api/","benchmark_results":[{"benchmark_name":"Arena-elo","benchmark_score":886},{"benchmark_name":"default","benchmark_score":886}],"deployment_information":{"datacenters":[{"country_code":"FR"}]},"model_base_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/v1","model_specs":{"capabilities":{"batch":false,"fill_in_the_middle":false,"function_calling":false,"input_modality":["text"],"output_modality":["image"],"reasoning":false,"response_format":[],"streaming":false,"use_case_tag":["image_generation"]},"model_parameters":{"parameters_type":"image_generation","params":[]},"nb_of_billion_parameters":3.5,"quantization":"fp32"},"playground_url":"https://playground.endpoints.ai.ovh.net/stable-diffusion-xl-base-v10","publishing_information":{"endpoint_publication_date":"24/10/2024","knowledge_cutoff":"None","languages":["*"],"licence_name":"OpenRail++","licence_url":"https://stable-diffusion-xl.endpoints.kepler.ai.cloud.ovh.net/doc/license","model_eol_date":"None","model_publication_date":"25/07/2023","publisher":"stabilityai","series":"stable-diffusion","source":"huggingface","source_link":"https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0"},"router_base_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/v1","tags":["Image Generation"],"usage_information":{"avg_throughtput":2,"avg_throughtput_unit":"empirical","pricing":[{"price":0,"price_unit":"per_image"}],"rate_limit":{"rate_limit_api":[{"rate_limit_unit":"req/m","rate_limit_value":400}],"rate_limit_api_anonymous":[{"rate_limit_unit":"req/m","rate_limit_value":2}],"rate_limit_playground":[{"rate_limit_unit":"req/m","rate_limit_value":2}]},"status":"live","statuses":["live"]}},"name":"stable-diffusion-xl-base-v10","new":false,"openapi_url":"https://stable-diffusion-xl.endpoints.kepler.ai.cloud.ovh.net/doc/openapi.json","router_openapi_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/doc/stable-diffusion-xl-base-v10/openapi.json","tags":["Image Generation"],"version":"v1.0"},{"available":true,"category":"Speech To Text","category_v2":"Speech To Text","created_at":"2025-08-12T09:49:44Z","description":"Whisper is a state-of-the-art model for automatic speech recognition (ASR) and speech translation, proposed in the paper \"Robust Speech Recognition via Large-Scale Weak Supervision\" by Alec Radford et al. from OpenAI. Trained on >5M hours of labeled data, Whisper demonstrates a strong ability to generalise to many datasets and domains in a zero-shot setting.\n","documentation_url":"https://whisper-large-v3.endpoints.kepler.ai.cloud.ovh.net/doc/readme","gradio_url":"https://playground.endpoints.ai.ovh.net/whisper-large-v3","id":"whisper-large-v3","isNew":false,"licence_url":"https://whisper-large-v3.endpoints.kepler.ai.cloud.ovh.net/doc/license","metadata":{"aliases":["openai/whisper-large-v3"],"api_base_url":"https://whisper-large-v3.endpoints.kepler.ai.cloud.ovh.net/api/openai_compat/","api_model_name":"whisper-large-v3","benchmark_results":[{"benchmark_name":"Average WER","benchmark_score":7.44},{"benchmark_name":"AMI","benchmark_score":15.95},{"benchmark_name":"Earnings22","benchmark_score":11.29},{"benchmark_name":"Gigaspeech","benchmark_score":10.02},{"benchmark_name":"LS Clean","benchmark_score":2.01},{"benchmark_name":"LS Other","benchmark_score":3.91},{"benchmark_name":"SPGISpeech","benchmark_score":2.94}],"deployment_information":{"datacenters":[{"country_code":"FR"}]},"model_base_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/v1","model_specs":{"capabilities":{"batch":false,"fill_in_the_middle":false,"function_calling":false,"input_modality":["sound"],"output_modality":["text"],"reasoning":false,"response_format":["json","verbose_json","text"],"streaming":false,"use_case_tag":["speech_to_text","diarisation","translation"]},"model_parameters":{"parameters_type":"asr","params":{"input_languages":["*"],"max_audio_duration":10800,"max_audio_duration_unauthenticated":60,"max_audio_size":2048,"max_audio_size_unauthenticated":10,"output_languages":["*"],"supported_formats":["mp3","mp4","aac","m4a","wav","flac","ogg","opus","webm","mpeg","mpga"]}},"nb_of_billion_parameters":1.54,"quantization":"fp16"},"playground_url":"https://playground.endpoints.ai.ovh.net/whisper-large-v3","publishing_information":{"endpoint_publication_date":"15/07/2025","knowledge_cutoff":"None","languages":["*"],"licence_name":"Apache 2.0","licence_url":"https://whisper-large-v3.endpoints.kepler.ai.cloud.ovh.net/doc/license","model_eol_date":"None","model_publication_date":"01/11/2023","publisher":"OpenAi","series":"Whisper","source":"huggingface","source_link":"https://huggingface.co/openai/whisper-large-v3"},"router_base_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/v1","usage_information":{"avg_throughtput":1,"avg_throughtput_unit":"empirical","pricing":[{"price":0.00004083,"price_unit":"audio_duration_seconds"}],"rate_limit":{"rate_limit_api":[{"rate_limit_unit":"req/m","rate_limit_value":400}],"rate_limit_api_anonymous":[{"rate_limit_unit":"req/m","rate_limit_value":2}],"rate_limit_playground":[{"rate_limit_unit":"req/m","rate_limit_value":2}]},"status":"live","statuses":["live"]}},"name":"whisper-large-v3","new":false,"openapi_url":"https://whisper-large-v3.endpoints.kepler.ai.cloud.ovh.net/doc/openapi.json","router_openapi_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/doc/whisper-large-v3/openapi.json","tags":["Automatic Speech Recognition","Diarisation","Translation"],"version":"v1.0"},{"available":true,"category":"Speech To Text","category_v2":"Speech To Text","created_at":"2025-08-12T09:49:44Z","description":"Whisper is a state-of-the-art model for automatic speech recognition (ASR) and speech translation, proposed in the paper \"Robust Speech Recognition via Large-Scale Weak Supervision\" by Alec Radford et al. from OpenAI. Trained on >5M hours of labeled data, Whisper demonstrates a strong ability to generalise to many datasets and domains in a zero-shot setting.\nWhisper large-v3-turbo is a finetuned version of a pruned Whisper large-v3. In other words, it's the exact same model, except that the number of decoding layers have reduced from 32 to 4. As a result, the model is way faster, at the expense of a minor quality degradation.\n","documentation_url":"https://whisper-large-v3-turbo.endpoints.kepler.ai.cloud.ovh.net/doc/readme","gradio_url":"https://playground.endpoints.ai.ovh.net/whisper-large-v3-turbo","id":"whisper-large-v3-turbo","isNew":false,"licence_url":"https://whisper-large-v3-turbo.endpoints.kepler.ai.cloud.ovh.net/doc/license","metadata":{"aliases":["openai/whisper-large-v3-turbo"],"api_base_url":"https://whisper-large-v3-turbo.endpoints.kepler.ai.cloud.ovh.net/api/openai_compat/","api_model_name":"whisper-large-v3-turbo","benchmark_results":[{"benchmark_name":"Average WER","benchmark_score":7.83},{"benchmark_name":"AMI","benchmark_score":16.13},{"benchmark_name":"Earnings22","benchmark_score":11.63},{"benchmark_name":"Gigaspeech","benchmark_score":10.14},{"benchmark_name":"LS Clean","benchmark_score":2.1},{"benchmark_name":"LS Other","benchmark_score":4.24},{"benchmark_name":"SPGISpeech","benchmark_score":2.97}],"deployment_information":{"datacenters":[{"country_code":"FR"}]},"model_base_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/v1","model_specs":{"capabilities":{"batch":false,"fill_in_the_middle":false,"function_calling":false,"input_modality":["sound"],"output_modality":["text"],"reasoning":false,"response_format":["json","verbose_json","text"],"streaming":false,"use_case_tag":["speech_to_text"]},"model_parameters":{"parameters_type":"asr","params":{"input_languages":["*"],"max_audio_duration":10800,"max_audio_duration_unauthenticated":60,"max_audio_size":2048,"max_audio_size_unauthenticated":10,"output_languages":["*"],"supported_formats":["mp3","mp4","aac","m4a","wav","flac","ogg","opus","webm","mpeg","mpga"]}},"nb_of_billion_parameters":0.81,"quantization":"fp16"},"playground_url":"https://playground.endpoints.ai.ovh.net/whisper-large-v3-turbo","publishing_information":{"endpoint_publication_date":"15/07/2025","knowledge_cutoff":"None","languages":["*"],"licence_name":"Apache 2.0","licence_url":"https://whisper-large-v3-turbo.endpoints.kepler.ai.cloud.ovh.net/doc/license","model_eol_date":"None","model_publication_date":"01/09/2024","publisher":"OpenAi","series":"Whisper","source":"huggingface","source_link":"https://huggingface.co/openai/whisper-large-v3-turbo"},"router_base_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/v1","usage_information":{"avg_throughtput":1,"avg_throughtput_unit":"empirical","pricing":[{"price":0.00001278,"price_unit":"audio_duration_seconds"}],"rate_limit":{"rate_limit_api":[{"rate_limit_unit":"req/m","rate_limit_value":400}],"rate_limit_api_anonymous":[{"rate_limit_unit":"req/m","rate_limit_value":2}],"rate_limit_playground":[{"rate_limit_unit":"req/m","rate_limit_value":2}]},"status":"live","statuses":["live"]}},"name":"whisper-large-v3-turbo","new":false,"openapi_url":"https://whisper-large-v3-turbo.endpoints.kepler.ai.cloud.ovh.net/doc/openapi.json","router_openapi_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/doc/whisper-large-v3-turbo/openapi.json","tags":["Automatic Speech Recognition","Diarisation","Translation"],"version":"v1.0"},{"available":true,"category":"Reasoning LLM","category_v2":"Text Generation","created_at":"2025-08-28T13:02:25Z","description":"gpt-oss-20b delivers strong performance with fast inference and efficient reasoning. Ideal for a wide range of tasks, it offers flexible deployment for developers seeking powerful, customizable AI solutions.\n","documentation_url":"https://gpt-oss-20b.endpoints.kepler.ai.cloud.ovh.net/doc/readme","gradio_url":"https://playground.endpoints.ai.ovh.net/gpt-oss-20b","id":"gpt-oss-20b","isNew":false,"licence_url":"https://gpt-oss-20b.endpoints.kepler.ai.cloud.ovh.net/doc/license","metadata":{"aliases":["openai/gpt-oss-20b"],"api_base_url":"https://gpt-oss-20b.endpoints.kepler.ai.cloud.ovh.net/api/openai_compat/","benchmark_results":[{"benchmark_name":"MMLU-Pro","benchmark_score":73.6},{"benchmark_name":"LiveCodeBench","benchmark_score":72.1}],"context":"131k","deployment_information":{"datacenters":[{"country_code":"FR"}]},"model_base_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/v1","model_specs":{"capabilities":{"batch":true,"fill_in_the_middle":false,"function_calling":true,"input_modality":["text"],"output_modality":["text"],"reasoning":true,"reasoning_efforts":["low","medium","high"],"response_format":["raw_text","json_object","json_schema"],"streaming":true,"use_case_tag":["code_chat","code_completion","code_explanation","emotion_detection","entity_extraction","multi_language_coding","multilingual","paraphrasing","sentiment_analysis","summarization","text_generation","text_only_compatible","translation"]},"model_parameters":{"parameters_type":"llm","params":{"context_size_k":131,"max_output_token_k":131}},"nb_of_billion_parameters":21,"quantization":"fp4"},"playground_url":"https://playground.endpoints.ai.ovh.net/gpt-oss-20b","publishing_information":{"endpoint_publication_date":"12/08/2025","knowledge_cutoff":"31/06/2024","languages":["*"],"licence_name":"Apache 2.0","licence_url":"https://gpt-oss-20b.endpoints.kepler.ai.cloud.ovh.net/doc/license","model_eol_date":"None","model_publication_date":"05/08/2025","publisher":"OpenAI","series":"gpt-oss","source":"huggingface","source_link":"https://huggingface.co/openai/gpt-oss-20b"},"router_base_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/v1","tags":["Reasoning","Streaming"],"usage_information":{"avg_throughtput":2,"avg_throughtput_unit":"empirical","pricing":[{"price":0.04,"price_unit":"million_input_tokens"},{"price":0.15,"price_unit":"million_output_tokens"}],"rate_limit":{"rate_limit_api":[{"rate_limit_unit":"req/m","rate_limit_value":400}],"rate_limit_api_anonymous":[{"rate_limit_unit":"req/m","rate_limit_value":2}],"rate_limit_playground":[{"rate_limit_unit":"req/m","rate_limit_value":2}]},"status":"live","statuses":["live"]}},"name":"gpt-oss-20b","new":false,"openapi_url":"https://gpt-oss-20b.endpoints.kepler.ai.cloud.ovh.net/doc/openapi.json","router_openapi_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/doc/gpt-oss-20b/openapi.json","tags":["Reasoning"],"version":"v1"},{"available":true,"category":"Text To Speech","category_v2":"Text To Speech","created_at":"2024-10-21T13:04:36Z","description":"This NVIDIA TTS model generates natural-sounding American English speech from raw text without requiring additional information.\n","documentation_url":"https://nvr-tts-en-us.endpoints.kepler.ai.cloud.ovh.net/doc/readme","gradio_url":"https://playground.endpoints.ai.ovh.net/nvr-tts-en-us","id":"nvr-tts-en-us","isNew":false,"licence_url":"https://nvr-tts-en-us.endpoints.kepler.ai.cloud.ovh.net/doc/license","metadata":{"aliases":["NVIDIA-RIVA"],"api_base_url":"https://nvr-tts-en-us.endpoints.kepler.ai.cloud.ovh.net/api/v1/","deployment_information":{"datacenters":[{"country_code":"FR"}]},"model_base_url":"https://nvr-tts-en-us.endpoints.kepler.ai.cloud.ovh.net/api/v1","model_specs":{"capabilities":{"batch":false,"fill_in_the_middle":false,"function_calling":false,"input_modality":["text"],"output_modality":["sound"],"reasoning":false,"response_format":["audio"],"streaming":false,"use_case_tag":["text-to-speech","multi-speaker"]},"model_parameters":{"parameters_type":"tts","params":{"input_languages":["en-us"],"output_languages":["en-us"]}},"quantization":"fp32"},"playground_url":"https://playground.endpoints.ai.ovh.net/nvr-tts-en-us","publishing_information":{"endpoint_publication_date":"28/06/2024","knowledge_cutoff":"None","languages":["en-us"],"licence_name":"Riva license","licence_url":"https://nvr-tts-en-us.endpoints.kepler.ai.cloud.ovh.net/doc/license","model_eol_date":"None","model_publication_date":"None","publisher":"NVIDIA Riva","series":"NVIDIA Riva","source":"NVIDIA"},"router_base_url":"https://oai.endpoints.kepler.ai.cloud.ovh.net/v1","usage_information":{"avg_throughtput":1,"avg_throughtput_unit":"empirical","pricing":[{"price":0,"price_unit":"million_input_chars"}],"rate_limit":{"rate_limit_api":[{"rate_limit_unit":"req/m","rate_limit_value":400}],"rate_limit_api_anonymous":[{"rate_limit_unit":"req/m","rate_limit_value":2}],"rate_limit_playground":[{"rate_limit_unit":"req/m","rate_limit_value":2}]},"status":"live","statuses":["live"]}},"name":"nvr-tts-en-us","new":false,"openapi_url":"https://nvr-tts-en-us.endpoints.kepler.ai.cloud.ovh.net/doc/openapi.json","router_openapi_url":"https://nvr-tts-en-us.endpoints.kepler.ai.cloud.ovh.net/doc/openapi.json","tags":["Text To Speech","Multi-speaker"],"version":"v1.0"}]