{"openapi":"3.0.0","info":{"description":"Featherless serverless AI inference API - OpenAI-compatible chat and text completions, embeddings, tokenization, text-to-speech with a live voice inventory, model catalog with per-model detail and plan/concurrency introspection; flat-rate Bearer API key.","version":"v1"},"servers":[{"description":"Featherless serverless AI inference API - OpenAI-compatible chat and text completions, embeddings, tokenization, text-to-speech with a live voice inventory, model catalog with per-model detail and plan/concurrency introspection; flat-rate Bearer API key.","url":"https://api.apirails.com/featherless"}],"paths":{"/v1/chat/completions":{"post":{"tags":["Chat Completions"],"parameters":[],"responses":{"401":{"description":"Bearer token required (Authorization header)","content":{"application/json":{}}}},"externalDocs":{"api_ref":"https://featherless.ai/docs/api-overview-and-common-options","api_endpoint":"https://api.featherless.ai/v1/chat/completions"},"summary":"Create chat completion","description":"Create chat completion | The chat endpoint allows callers to specify the prompt in terms of an exchange between a user and an assistant. | Source: official API docs at https://featherless.ai/docs/","security":[{"bearerAuth":[]}],"requestBody":{"description":"","required":true,"content":{"application/json":{"schema":{"type":"object","required":[],"properties":{"chat_template_kwargs":{"type":"object","description":"Optional model-specific chat template parameters, such as toggling reasoning (enable_thinking, thinking, do_reasoning, thinking_budget) - see the chat_template_kwargs guide.","properties":{"enable_thinking":{"type":"boolean","description":"Request thinking/reasoning behavior for models that support it (synonyms: thinking, do_reasoning).","enum":[true,false]},"thinking_budget":{"type":"number","description":"Reasoning token budget, for templates that support it."}}},"frequency_penalty":{"type":"number","description":"Penalizes new tokens based on their frequency in the generated text so far. Values greater than 0 encourage new tokens; less than 0 encourages repetition."},"include_stop_str_in_output":{"type":"boolean","description":"Boolean to include stop strings in output text. Defaults to false.","enum":[true,false]},"max_tokens":{"type":"number","description":"Maximum number of tokens generated per output sequence."},"messages":{"type":"array","description":"A list of messages comprising the conversation so far (OpenAI-compatible message objects with role and content).","required":true,"items":{"type":"object","properties":{"content":{"type":"string","description":"Message content (string; multimodal models also accept structured content parts).","required":true},"role":{"type":"string","description":"Conversation role (system, user, assistant or tool).","required":true}}}},"min_p":{"type":"number","description":"Minimum probability for a token to be considered relative to the most likely token. Must be in [0, 1]. Set to 0 to disable."},"min_tokens":{"type":"number","description":"Minimum number of tokens generated per output sequence before EOS or stop_token_ids can be generated."},"model":{"type":"string","description":"ID of the model to use, from the model catalog (https://featherless.ai/models), e.g. meta-llama/Meta-Llama-3.1-8B-Instruct.","required":true},"presence_penalty":{"type":"number","description":"Penalizes new tokens based on their presence in the generated text so far. Values greater than 0 encourage new tokens; less than 0 encourages repetition."},"repetition_penalty":{"type":"number","description":"Penalizes new tokens based on their appearance in the prompt and generated text. Values greater than 1 encourage new tokens; less than 1 encourages repetition."},"seed":{"type":"number","description":"Random seed for generation. Not reliable, as multiple servers may be used."},"stop":{"type":"array","description":"List of strings that stop generation when generated. The returned output excludes these strings.","items":{"type":"string"}},"stop_token_ids":{"type":"array","description":"List of token ids that stop generation when generated. The returned output may include these tokens unless they are special tokens.","items":{"type":"number"}},"temperature":{"type":"number","description":"Controls sampling randomness. Lower values make the model more deterministic; higher values introduce randomness. Zero is greedy sampling."},"top_k":{"type":"number","description":"Number of top tokens to consider. Set to -1 to consider all tokens."},"top_p":{"type":"number","description":"Controls the cumulative probability of considered top tokens. Must be in (0, 1]. Set to 1 to consider all tokens."}}},"example":{"chat_template_kwargs":[{"enable_thinking":false}],"frequency_penalty":[0],"include_stop_str_in_output":[false],"max_tokens":[512],"messages":[[{"content":"Hello!","role":"user"}]],"min_p":[0],"min_tokens":[1],"model":["meta-llama/Meta-Llama-3.1-8B-Instruct"],"presence_penalty":[0],"repetition_penalty":[1],"seed":[42],"stop":[[["\n\nUser:"]]],"stop_token_ids":[[[2]]],"temperature":[0.7],"top_k":[-1],"top_p":[1]}}}}}},"/v1/completions":{"post":{"tags":["Text Completions"],"parameters":[],"responses":{"401":{"description":"Bearer token required (Authorization header)","content":{"application/json":{}}}},"externalDocs":{"api_ref":"https://featherless.ai/docs/completions","api_endpoint":"https://api.featherless.ai/v1/completions"},"summary":"Create text completion","description":"Create text completion | The text completions endpoint takes a single string (the prompt) and generates a completion, for fine-grained control of chat format or non-assistant scenarios. | Source: official API docs at https://featherless.ai/docs/","security":[{"bearerAuth":[]}],"requestBody":{"description":"","required":true,"content":{"application/json":{"schema":{"type":"object","required":[],"properties":{"frequency_penalty":{"type":"number","description":"Penalizes new tokens based on their frequency in the generated text so far. Values greater than 0 encourage new tokens; less than 0 encourages repetition."},"include_stop_str_in_output":{"type":"boolean","description":"If set to true, includes stop strings in the output text. Defaults to false.","enum":[true,false]},"max_tokens":{"type":"number","description":"Maximum number of tokens generated per output sequence."},"min_p":{"type":"number","description":"Minimum probability for a token to be considered relative to the most likely token. Must be in [0, 1]. Set to 0 to disable."},"min_tokens":{"type":"number","description":"Minimum number of tokens generated per output sequence before EOS or stop_token_ids can be generated."},"model":{"type":"string","description":"ID of the model to use, from the model catalog (https://featherless.ai/models), e.g. meta-llama/Meta-Llama-3.1-8B-Instruct.","required":true},"presence_penalty":{"type":"number","description":"Penalizes new tokens based on their presence in the generated text so far. Values greater than 0 encourage new tokens; less than 0 encourages repetition."},"prompt":{"type":"string","description":"The prompt(s) to generate completions for, encoded as a string. An array of strings is also accepted and is treated as a parallel inference call subject to the same concurrency restrictions as parallel independent calls.","required":true},"repetition_penalty":{"type":"number","description":"Penalizes new tokens based on their appearance in the prompt and generated text. Values greater than 1 encourage new tokens; less than 1 encourages repetition."},"seed":{"type":"number","description":"Random seed for generation. Not reliable, as multiple servers may be used."},"stop":{"type":"array","description":"List of strings that stop generation when generated. The returned output excludes these strings.","items":{"type":"string"}},"stop_token_ids":{"type":"array","description":"List of token ids that stop generation when generated. The returned output may include these tokens unless they are special tokens.","items":{"type":"number"}},"temperature":{"type":"number","description":"Controls sampling randomness. Lower values make the model more deterministic; higher values introduce randomness. Zero is greedy sampling."},"top_k":{"type":"number","description":"Number of top tokens to consider. Set to -1 to consider all tokens."},"top_p":{"type":"number","description":"Controls the cumulative probability of considered top tokens. Must be in (0, 1]. Set to 1 to consider all tokens."}}},"example":{"frequency_penalty":[0],"include_stop_str_in_output":[false],"max_tokens":[512],"min_p":[0],"min_tokens":[1],"model":["meta-llama/Meta-Llama-3.1-8B-Instruct"],"presence_penalty":[0],"prompt":["Once upon a time,"],"repetition_penalty":[1],"seed":[42],"stop":[[["\n\nUser:"]]],"stop_token_ids":[[[2]]],"temperature":[0.7],"top_k":[-1],"top_p":[1]}}}}}},"/v1/embeddings":{"post":{"tags":["Embeddings"],"parameters":[],"responses":{"401":{"description":"Bearer token required (Authorization header)","content":{"application/json":{}}}},"externalDocs":{"api_ref":"https://featherless.ai/docs/embeddings","api_endpoint":"https://api.featherless.ai/v1/embeddings"},"summary":"Create embedding","description":"Create embedding | OpenAI-compatible embeddings requests for models that produce embedding outputs; usable for search, similarity, clustering, classification and retrieval pipelines. | Source: official API docs at https://featherless.ai/docs/","security":[{"bearerAuth":[]}],"requestBody":{"description":"","required":true,"content":{"application/json":{"schema":{"type":"object","required":[],"properties":{"dimensions":{"type":"number","description":"Optional output dimensionality. Use only if your chosen model supports it."},"encoding_format":{"type":"string","description":"Optional encoding format of the embedding output. Usually float."},"input":{"type":"string","description":"A string or array of strings to embed. Multiple strings return one embedding per input in the same order.","required":true},"model":{"type":"string","description":"The embedding model to use; supported models are surfaced in the catalog at https://featherless.ai/models?modalities=embedding.","required":true}}},"example":{"dimensions":[1024],"encoding_format":["float"],"input":["The quick brown fox jumps over the lazy dog"],"model":["BAAI/bge-m3"]}}}}}},"/v1/tokenize":{"post":{"tags":["Tokenizer"],"parameters":[],"responses":{"401":{"description":"Bearer token required (Authorization header)","content":{"application/json":{}}}},"externalDocs":{"api_ref":"https://featherless.ai/docs/v1-tokenize","api_endpoint":"https://api.featherless.ai/v1/tokenize"},"summary":"Tokenize text","description":"Tokenize text | Applies a specific model tokenizer to an input string; useful to obtain token counts post-hoc for streaming requests. | Source: official API docs at https://featherless.ai/docs/","security":[{"bearerAuth":[]}],"requestBody":{"description":"","required":true,"content":{"application/json":{"schema":{"type":"object","required":[],"properties":{"text":{"type":"string","description":"The string to apply the tokenizer to.","required":true},"model":{"type":"string","description":"ID of the model to use, from the model catalog (https://featherless.ai/models), e.g. meta-llama/Meta-Llama-3.1-8B-Instruct.","required":true}}},"example":{"text":["Hello, world!"],"model":["meta-llama/Meta-Llama-3.1-8B-Instruct"]}}}}}},"/v1/audio/speech":{"post":{"tags":["Text To Speech"],"parameters":[],"responses":{"401":{"description":"Bearer token required (Authorization header)","content":{"application/json":{}}}},"externalDocs":{"api_ref":"https://featherless.ai/docs/api-reference-audio-speech","api_endpoint":"https://api.featherless.ai/v1/audio/speech"},"summary":"Generate speech audio","description":"Generate speech audio | Text-to-speech generation over the speech model catalog; billed on input characters. Successful responses carry X-Generation-Id, X-Request-Id and X-Input-Characters headers. | Source: official API docs at https://featherless.ai/docs/","security":[{"bearerAuth":[]}],"requestBody":{"description":"","required":true,"content":{"application/json":{"schema":{"type":"object","required":[],"properties":{"delivery":{"type":"string","description":"Delivery control: bulk (raw binary, default), json (base64 envelope), or stream (SSE base64 events).","enum":["bulk","json","stream"]},"encoding":{"type":"string","description":"Delivery encoding: binary (bulk default) or base64.","enum":["binary","base64"]},"input":{"type":"string","description":"Target text, forwarded verbatim. Each model applies its own length limit; expression and speaker syntax, when supported, is written directly into this string. Maximum request body is 50 MiB including base64 reference audio.","required":true},"instructions":{"type":"string","description":"Optional speaking-style guidance. Not guaranteed to affect the output for a given model; only use when the model guide marks it as exposed."},"model":{"type":"string","description":"The exact speech-model ID returned by the model catalog (e.g. hexgrad/Kokoro-82M, ResembleAI/chatterbox-turbo); list them with model_class=tts on GET /v1/models.","required":true},"response_format":{"type":"string","description":"Requested audio container. A request, not a guarantee: only PCM-to-WAV rewrapping is performed, no codec transcoding - inspect the response Content-Type.","enum":["mp3","opus","aac","flac","wav","pcm"]},"speed":{"type":"number","description":"Speech speed from 0.25 through 4; values outside the range fail validation. Defaults to 1."},"stream":{"type":"boolean","description":"Delivery control: true selects SSE streaming. Precedence: delivery, then stream_format, then stream.","enum":[true,false]},"stream_format":{"type":"string","description":"Delivery control alias: sse selects SSE, audio selects bulk audio. Precedence: delivery, then stream_format, then stream.","enum":["sse","audio"]},"voice":{"type":"string","description":"Optional voice selection: a preset name string from GET /v1/audio/voices (e.g. af_bella), \"default\", a custom ID object, or a voice-cloning object (voice.clone.audio.data = bare base64 clip, voice.clone.audio.format = wav or mp3)."}}},"example":{"delivery":["bulk"],"encoding":["base64"],"input":["Hello from Featherless!"],"instructions":["Speak cheerfully"],"model":["hexgrad/Kokoro-82M"],"response_format":["mp3"],"speed":[1],"stream":[false],"stream_format":["sse"],"voice":["af_bella"]}}}}}},"/v1/audio/voices":{"get":{"tags":["Text To Speech"],"parameters":[{"name":"model","in":"query","description":"Model ID whose voice inventory to list, e.g. hexgrad/Kokoro-82M. The inventory is the authority on available voices - a name it does not list is refused with 400 invalid_voice.","required":true,"schema":{"type":"string","example":""}}],"responses":{"401":{"description":"Bearer token required (Authorization header)","content":{"application/json":{}}}},"externalDocs":{"api_ref":"https://featherless.ai/docs/audio-speech","api_endpoint":"https://api.featherless.ai/v1/audio/voices"},"summary":"List speech voice inventory","description":"List speech voice inventory | Returns the live preset voice inventory for a speech model. Voice IDs are case-sensitive; voice prefixes encode language and voice group. | Source: official API docs at https://featherless.ai/docs/","security":[{"bearerAuth":[]}]}},"/v1/models":{"get":{"tags":["Models"],"parameters":[{"name":"architectures","in":"query","description":"Filter by model architectures. Multi-value filters can be passed as comma-separated values or repeated query parameters.","required":false,"schema":{"type":"string","example":""}},{"name":"available_on_current_plan","in":"query","description":"Pass true or 1 to filter to models available on the authenticated user current plan.","required":false,"schema":{"type":"boolean","example":"","items":{"type":"string","enum":[true,false]}}},{"name":"capabilities","in":"query","description":"Filter by capabilities, e.g. chat,tool-use. Multi-value filters can be passed as comma-separated values or repeated query parameters.","required":false,"schema":{"type":"array","example":""}},{"name":"content_flags","in":"query","description":"Filter by content flags. Multi-value filters can be passed as comma-separated values or repeated query parameters.","required":false,"schema":{"type":"string","example":""}},{"name":"context_length_max","in":"query","description":"Range filter: maximum context size in tokens.","required":false,"schema":{"type":"number","example":""}},{"name":"context_length_min","in":"query","description":"Range filter: minimum context size in tokens.","required":false,"schema":{"type":"number","example":""}},{"name":"conversational","in":"query","description":"Filter by conversational support: true or false.","required":false,"schema":{"type":"boolean","example":"","items":{"type":"string","enum":[true,false]}}},{"name":"creative","in":"query","description":"Filter by creative model flag. Multi-value filters can be passed as comma-separated values or repeated query parameters.","required":false,"schema":{"type":"string","example":""}},{"name":"domains","in":"query","description":"Filter by domains. Multi-value filters can be passed as comma-separated values or repeated query parameters.","required":false,"schema":{"type":"string","example":""}},{"name":"family","in":"query","description":"Filter by model family, e.g. llama3, qwen2. Multi-value filters can be passed as comma-separated values or repeated query parameters.","required":false,"schema":{"type":"array","example":""}},{"name":"gated","in":"query","description":"Filter by Hugging Face gated status: true or false.","required":false,"schema":{"type":"boolean","example":"","items":{"type":"string","enum":[true,false]}}},{"name":"languages","in":"query","description":"Filter by language codes, e.g. en,ja. Multi-value filters can be passed as comma-separated values or repeated query parameters.","required":false,"schema":{"type":"array","example":""}},{"name":"license","in":"query","description":"Filter by license. Multi-value filters can be passed as comma-separated values or repeated query parameters.","required":false,"schema":{"type":"string","example":""}},{"name":"modalities","in":"query","description":"Filter by modalities, e.g. vision or embedding. Multi-value filters can be passed as comma-separated values or repeated query parameters.","required":false,"schema":{"type":"array","example":""}},{"name":"model_class","in":"query","description":"Filter by model class, e.g. chat models or model_class=tts for speech models. Multi-value filters can be passed as comma-separated values or repeated query parameters.","required":false,"schema":{"type":"string","example":""}},{"name":"page","in":"query","description":"Page number for pagination.","required":false,"schema":{"type":"number","example":""}},{"name":"parameter_bucket","in":"query","description":"Filter by parameter bucket. Multi-value filters can be passed as comma-separated values or repeated query parameters.","required":false,"schema":{"type":"string","example":""}},{"name":"per_page","in":"query","description":"Number of results per page. Maximum is 1000.","required":false,"schema":{"type":"number","example":""}},{"name":"popularity_level","in":"query","description":"Filter by popularity level. Multi-value filters can be passed as comma-separated values or repeated query parameters.","required":false,"schema":{"type":"string","example":""}},{"name":"q","in":"query","description":"Search by model name or id. Multi-value filters can be passed as comma-separated values or repeated query parameters.","required":false,"schema":{"type":"string","example":""}},{"name":"status","in":"query","description":"Deployment status: active, pending_deploy or not_deployed. When omitted, only active models are returned. Multi-value filters can be passed as comma-separated values or repeated query parameters.","required":false,"schema":{"type":"array","example":""}},{"name":"tags","in":"query","description":"Require all listed tags to match. This is a legacy filter. Multi-value filters can be passed as comma-separated values or repeated query parameters.","required":false,"schema":{"type":"array","example":""}},{"name":"tasks","in":"query","description":"Filter by pipeline tasks. Multi-value filters can be passed as comma-separated values or repeated query parameters.","required":false,"schema":{"type":"string","example":""}},{"name":"training","in":"query","description":"Filter by training provenance. Multi-value filters can be passed as comma-separated values or repeated query parameters.","required":false,"schema":{"type":"string","example":""}}],"responses":{"401":{"description":"Bearer token required (Authorization header)","content":{"application/json":{}}}},"externalDocs":{"api_ref":"https://featherless.ai/docs/api-reference-models","api_endpoint":"https://api.featherless.ai/v1/models"},"summary":"List models","description":"List models | Lists all models in the catalogue with current state; callable authenticated or unauthenticated. Use the id field as the model value when calling the completion endpoints. | Source: official API docs at https://featherless.ai/docs/","security":[{"bearerAuth":[]}]}},"/v1/models/{model-id}":{"get":{"tags":["Models"],"parameters":[{"name":"model-id","in":"path","description":"Model id. Ids contain a slash (owner/model): both the raw form and the URL-encoded single-segment form (owner%2Fmodel) are accepted - prefer the URL-encoded form here.","required":true,"schema":{"type":"string","example":""}}],"responses":{"401":{"description":"Bearer token required (Authorization header)","content":{"application/json":{}}}},"externalDocs":{"api_ref":"https://featherless.ai/docs/api-reference-models","api_endpoint":"https://api.featherless.ai/v1/models/{model-id}"},"summary":"Get model detail","description":"Get model detail | Fetch a single model by id: a superset of a list entry adding live hot/cold availability, deployment status, per-token pricing, concurrency cost and HuggingFace metadata. Unknown ids return 404 model_not_found. | Source: official API docs at https://featherless.ai/docs/","security":[{"bearerAuth":[]}]}},"/v1/plan":{"get":{"tags":["Plan"],"parameters":[],"responses":{"401":{"description":"Bearer token required (Authorization header)","content":{"application/json":{}}}},"externalDocs":{"api_ref":"https://featherless.ai/docs/api-reference-plan","api_endpoint":"https://api.featherless.ai/v1/plan"},"summary":"Get subscription plan","description":"Get subscription plan | Returns the plan the API key is subscribed to: max_context_length (null = no plan cap), max_model_size (null = no limit) and concurrency. Requires authentication. | Source: official API docs at https://featherless.ai/docs/","security":[{"bearerAuth":[]}]}},"/account/concurrency":{"get":{"tags":["Account"],"parameters":[],"responses":{"401":{"description":"Bearer token required (Authorization header)","content":{"application/json":{}}}},"externalDocs":{"api_ref":"https://featherless.ai/docs/api-reference-concurrency-stream","api_endpoint":"https://api.featherless.ai/account/concurrency"},"summary":"Get concurrency usage snapshot","description":"Get concurrency usage snapshot | One-off polling snapshot of in-flight requests against the organization concurrency allotment; same payload as a single SSE frame, as regular JSON. | Source: official API docs at https://featherless.ai/docs/","security":[{"bearerAuth":[]}]}},"/account/concurrency/stream":{"get":{"tags":["Account"],"parameters":[],"responses":{"401":{"description":"Bearer token required (Authorization header)","content":{"application/json":{}}}},"externalDocs":{"api_ref":"https://featherless.ai/docs/api-reference-concurrency-stream","api_endpoint":"https://api.featherless.ai/account/concurrency/stream"},"summary":"Stream concurrency usage","description":"Stream concurrency usage | Server-Sent Events stream of live concurrency usage: one event immediately on connect and then one every two seconds, until the client disconnects. | Source: official API docs at https://featherless.ai/docs/","security":[{"bearerAuth":[]}]}},"/debug/chat-format":{"post":{"tags":["Chat Completions"],"parameters":[],"responses":{"401":{"description":"Bearer token required (Authorization header)","content":{"application/json":{}}}},"externalDocs":{"api_ref":"https://featherless.ai/docs/chat-template-kwargs","api_endpoint":"https://api.featherless.ai/debug/chat-format"},"summary":"Preview chat format","description":"Preview chat format | Debug endpoint: previews how a chat request will be formatted. Response includes formatted_prompt, token_count and template_info. | Source: official API docs at https://featherless.ai/docs/","security":[{"bearerAuth":[]}],"requestBody":{"description":"","required":true,"content":{"application/json":{"schema":{"type":"object","required":[],"properties":{"chat_template_kwargs":{"type":"object","description":"Model-specific chat template parameters to include in the rendered prompt (e.g. enable_thinking, thinking_budget).","properties":{}},"messages":{"type":"array","description":"The conversation messages to render through the model chat template (chat/completions request shape).","required":true,"items":{"type":"object","properties":{"content":{"type":"string","description":"Message content.","required":true},"role":{"type":"string","description":"Conversation role (system, user, assistant or tool).","required":true}}}},"model":{"type":"string","description":"ID of the model to use, from the model catalog (https://featherless.ai/models), e.g. meta-llama/Meta-Llama-3.1-8B-Instruct.","required":true}}},"example":{"chat_template_kwargs":[{"enable_thinking":false}],"messages":[[{"content":"Hello!","role":"user"}]],"model":["meta-llama/Meta-Llama-3.1-8B-Instruct"]}}}}}},"/models/{owner}/{model}/debug/chat-format":{"post":{"tags":["Chat Completions"],"parameters":[{"name":"model","in":"path","description":"Model name within the owner namespace, e.g. Meta-Llama-3.1-8B-Instruct.","required":true,"schema":{"type":"string","example":""}},{"name":"owner","in":"path","description":"Model owner namespace, e.g. meta-llama.","required":true,"schema":{"type":"string","example":""}}],"responses":{"401":{"description":"Bearer token required (Authorization header)","content":{"application/json":{}}}},"externalDocs":{"api_ref":"https://featherless.ai/docs/chat-template-kwargs","api_endpoint":"https://api.featherless.ai/models/{owner}/{model}/debug/chat-format"},"summary":"Preview model chat format","description":"Preview model chat format | Debug endpoint scoped to a specific model: previews how its chat request will be formatted. Response includes formatted_prompt, token_count and template_info. | Source: official API docs at https://featherless.ai/docs/","security":[{"bearerAuth":[]}],"requestBody":{"description":"","required":true,"content":{"application/json":{"schema":{"type":"object","required":[],"properties":{"chat_template_kwargs":{"type":"object","description":"Model-specific chat template parameters to include in the rendered prompt (e.g. enable_thinking, thinking_budget).","properties":{}},"messages":{"type":"array","description":"The conversation messages to render through the model chat template (chat/completions request shape).","required":true,"items":{"type":"object","properties":{"content":{"type":"string","description":"Message content.","required":true},"role":{"type":"string","description":"Conversation role (system, user, assistant or tool).","required":true}}}},"model":{"type":"string","description":"ID of the model to use, from the model catalog (https://featherless.ai/models), e.g. meta-llama/Meta-Llama-3.1-8B-Instruct.","required":true}}},"example":{"chat_template_kwargs":[{"enable_thinking":false}],"messages":[[{"content":"Hello!","role":"user"}]],"model":["meta-llama/Meta-Llama-3.1-8B-Instruct"]}}}}}}},"components":{"securitySchemes":{"bearerAuth":{"type":"http","scheme":"bearer"}}},"security":[{"bearerAuth":[]}]}