@huggingface/inference 3.1.6 → 3.3.0

package/README.md

@@ -46,7 +46,12 @@ Your access token should be kept private. If you need to protect it in front-end
 
 You can send inference requests to third-party providers with the inference client.
 
-Currently, we support the following providers: [Fal.ai](https://fal.ai), [Replicate](https://replicate.com), [Together](https://together.xyz) and [Sambanova](https://sambanova.ai).
+Currently, we support the following providers:
+- [Fal.ai](https://fal.ai)
+- [Fireworks AI](https://fireworks.ai)
+- [Replicate](https://replicate.com)
+- [Sambanova](https://sambanova.ai)
+- [Together](https://together.xyz)
 
 To send requests to a third-party provider, you have to pass the `provider` parameter to the inference function. Make sure your request is authenticated with an access token.
 ```ts
@@ -64,10 +69,11 @@ When authenticated with a Hugging Face access token, the request is routed throu
 When authenticated with a third-party provider key, the request is made directly against that provider's inference API.
 
 Only a subset of models are supported when requesting third-party providers. You can check the list of supported models per pipeline tasks here:
-- [Fal.ai supported models](./src/providers/fal-ai.ts)
-- [Replicate supported models](./src/providers/replicate.ts)
-- [Sambanova supported models](./src/providers/sambanova.ts)
-- [Together supported models](./src/providers/together.ts)
+- [Fal.ai supported models](https://huggingface.co/api/partners/fal-ai/models)
+- [Fireworks AI supported models](https://huggingface.co/api/partners/fireworks-ai/models)
+- [Replicate supported models](https://huggingface.co/api/partners/replicate/models)
+- [Sambanova supported models](https://huggingface.co/api/partners/sambanova/models)
+- [Together supported models](https://huggingface.co/api/partners/together/models)
 - [HF Inference API (serverless)](https://huggingface.co/models?inference=warm&sort=trending)
 
 ❗**Important note:** To be compatible, the third-party API must adhere to the "standard" shape API we expect on HF model pages for each pipeline task type. 
@@ -117,7 +123,7 @@ for await (const output of hf.textGenerationStream({
 
 ### Text Generation (Chat Completion API Compatible)
 
-Using the `chatCompletion` method, you can generate text with models compatible with the OpenAI Chat Completion API. All models served by [TGI](https://api-inference.huggingface.co/framework/text-generation-inference) on Hugging Face support Messages API.
+Using the `chatCompletion` method, you can generate text with models compatible with the OpenAI Chat Completion API. All models served by [TGI](https://huggingface.co/docs/text-generation-inference/) on Hugging Face support Messages API.
 
 [Demo](https://huggingface.co/spaces/huggingfacejs/streaming-chat-completion)
 
@@ -611,7 +617,7 @@ const { generated_text } = await gpt2.textGeneration({inputs: 'The answer to the
 
 // Chat Completion Example
 const ep = hf.endpoint(
-  "https://api-inference.huggingface.co/models/meta-llama/Llama-3.1-8B-Instruct"
+  "https://router.huggingface.co/hf-inference/models/meta-llama/Llama-3.1-8B-Instruct"
 );
 const stream = ep.chatCompletionStream({
   model: "tgi",

package/dist/index.cjs

@@ -20,14 +20,10 @@ var __toCommonJS = (mod) => __copyProps(__defProp({}, "__esModule", { value: tru
 // src/index.ts
 var src_exports = {};
 __export(src_exports, {
-  FAL_AI_SUPPORTED_MODEL_IDS: () => FAL_AI_SUPPORTED_MODEL_IDS,
   HfInference: () => HfInference,
   HfInferenceEndpoint: () => HfInferenceEndpoint,
   INFERENCE_PROVIDERS: () => INFERENCE_PROVIDERS,
   InferenceOutputError: () => InferenceOutputError,
-  REPLICATE_SUPPORTED_MODEL_IDS: () => REPLICATE_SUPPORTED_MODEL_IDS,
-  SAMBANOVA_SUPPORTED_MODEL_IDS: () => SAMBANOVA_SUPPORTED_MODEL_IDS,
-  TOGETHER_SUPPORTED_MODEL_IDS: () => TOGETHER_SUPPORTED_MODEL_IDS,
   audioClassification: () => audioClassification,
   audioToAudio: () => audioToAudio,
   automaticSpeechRecognition: () => automaticSpeechRecognition,
@@ -102,128 +98,22 @@ __export(tasks_exports, {
 
 // src/config.ts
 var HF_HUB_URL = "https://huggingface.co";
+var HF_ROUTER_URL = "https://router.huggingface.co";
 
 // src/providers/fal-ai.ts
 var FAL_AI_API_BASE_URL = "https://fal.run";
-var FAL_AI_SUPPORTED_MODEL_IDS = {
-  "text-to-image": {
-    "black-forest-labs/FLUX.1-schnell": "fal-ai/flux/schnell",
-    "black-forest-labs/FLUX.1-dev": "fal-ai/flux/dev",
-    "playgroundai/playground-v2.5-1024px-aesthetic": "fal-ai/playground-v25",
-    "ByteDance/SDXL-Lightning": "fal-ai/lightning-models",
-    "PixArt-alpha/PixArt-Sigma-XL-2-1024-MS": "fal-ai/pixart-sigma",
-    "stabilityai/stable-diffusion-3-medium": "fal-ai/stable-diffusion-v3-medium",
-    "Warlord-K/Sana-1024": "fal-ai/sana",
-    "fal/AuraFlow-v0.2": "fal-ai/aura-flow",
-    "stabilityai/stable-diffusion-3.5-large": "fal-ai/stable-diffusion-v35-large",
-    "stabilityai/stable-diffusion-3.5-large-turbo": "fal-ai/stable-diffusion-v35-large/turbo",
-    "stabilityai/stable-diffusion-3.5-medium": "fal-ai/stable-diffusion-v35-medium",
-    "Kwai-Kolors/Kolors": "fal-ai/kolors"
-  },
-  "automatic-speech-recognition": {
-    "openai/whisper-large-v3": "fal-ai/whisper"
-  },
-  "text-to-video": {
-    "genmo/mochi-1-preview": "fal-ai/mochi-v1",
-    "tencent/HunyuanVideo": "fal-ai/hunyuan-video",
-    "THUDM/CogVideoX-5b": "fal-ai/cogvideox-5b",
-    "Lightricks/LTX-Video": "fal-ai/ltx-video"
-  }
-};
 
 // src/providers/replicate.ts
 var REPLICATE_API_BASE_URL = "https://api.replicate.com";
-var REPLICATE_SUPPORTED_MODEL_IDS = {
-  "text-to-image": {
-    "black-forest-labs/FLUX.1-dev": "black-forest-labs/flux-dev",
-    "black-forest-labs/FLUX.1-schnell": "black-forest-labs/flux-schnell",
-    "ByteDance/Hyper-SD": "bytedance/hyper-flux-16step:382cf8959fb0f0d665b26e7e80b8d6dc3faaef1510f14ce017e8c732bb3d1eb7",
-    "ByteDance/SDXL-Lightning": "bytedance/sdxl-lightning-4step:5599ed30703defd1d160a25a63321b4dec97101d98b4674bcc56e41f62f35637",
-    "playgroundai/playground-v2.5-1024px-aesthetic": "playgroundai/playground-v2.5-1024px-aesthetic:a45f82a1382bed5c7aeb861dac7c7d191b0fdf74d8d57c4a0e6ed7d4d0bf7d24",
-    "stabilityai/stable-diffusion-3.5-large-turbo": "stability-ai/stable-diffusion-3.5-large-turbo",
-    "stabilityai/stable-diffusion-3.5-large": "stability-ai/stable-diffusion-3.5-large",
-    "stabilityai/stable-diffusion-3.5-medium": "stability-ai/stable-diffusion-3.5-medium",
-    "stabilityai/stable-diffusion-xl-base-1.0": "stability-ai/sdxl:7762fd07cf82c948538e41f63f77d685e02b063e37e496e96eefd46c929f9bdc"
-  },
-  "text-to-speech": {
-    "OuteAI/OuteTTS-0.3-500M": "jbilcke/oute-tts:3c645149db020c85d080e2f8cfe482a0e68189a922cde964fa9e80fb179191f3",
-    "hexgrad/Kokoro-82M": "jaaari/kokoro-82m:dfdf537ba482b029e0a761699e6f55e9162cfd159270bfe0e44857caa5f275a6"
-  },
-  "text-to-video": {
-    "genmo/mochi-1-preview": "genmoai/mochi-1:1944af04d098ef69bed7f9d335d102e652203f268ec4aaa2d836f6217217e460"
-  }
-};
 
 // src/providers/sambanova.ts
 var SAMBANOVA_API_BASE_URL = "https://api.sambanova.ai";
-var SAMBANOVA_SUPPORTED_MODEL_IDS = {
-  /** Chat completion / conversational */
-  conversational: {
-    "deepseek-ai/DeepSeek-Distill-R1-Llama-70B": "DeepSeek-Distill-R1-Llama-70B",
-    "Qwen/Qwen2.5-Coder-32B-Instruct": "Qwen2.5-Coder-32B-Instruct",
-    "Qwen/Qwen2.5-72B-Instruct": "Qwen2.5-72B-Instruct",
-    "Qwen/QwQ-32B-Preview": "QwQ-32B-Preview",
-    "meta-llama/Llama-3.3-70B-Instruct": "Meta-Llama-3.3-70B-Instruct",
-    "meta-llama/Llama-3.2-1B-Instruct": "Meta-Llama-3.2-1B-Instruct",
-    "meta-llama/Llama-3.2-3B-Instruct": "Meta-Llama-3.2-3B-Instruct",
-    "meta-llama/Llama-3.2-11B-Vision-Instruct": "Llama-3.2-11B-Vision-Instruct",
-    "meta-llama/Llama-3.2-90B-Vision-Instruct": "Llama-3.2-90B-Vision-Instruct",
-    "meta-llama/Llama-3.1-8B-Instruct": "Meta-Llama-3.1-8B-Instruct",
-    "meta-llama/Llama-3.1-70B-Instruct": "Meta-Llama-3.1-70B-Instruct",
-    "meta-llama/Llama-3.1-405B-Instruct": "Meta-Llama-3.1-405B-Instruct",
-    "meta-llama/Llama-Guard-3-8B": "Meta-Llama-Guard-3-8B"
-  }
-};
 
 // src/providers/together.ts
 var TOGETHER_API_BASE_URL = "https://api.together.xyz";
-var TOGETHER_SUPPORTED_MODEL_IDS = {
-  "text-to-image": {
-    "black-forest-labs/FLUX.1-Canny-dev": "black-forest-labs/FLUX.1-canny",
-    "black-forest-labs/FLUX.1-Depth-dev": "black-forest-labs/FLUX.1-depth",
-    "black-forest-labs/FLUX.1-dev": "black-forest-labs/FLUX.1-dev",
-    "black-forest-labs/FLUX.1-Redux-dev": "black-forest-labs/FLUX.1-redux",
-    "black-forest-labs/FLUX.1-schnell": "black-forest-labs/FLUX.1-pro",
-    "stabilityai/stable-diffusion-xl-base-1.0": "stabilityai/stable-diffusion-xl-base-1.0"
-  },
-  conversational: {
-    "databricks/dbrx-instruct": "databricks/dbrx-instruct",
-    "deepseek-ai/DeepSeek-R1": "deepseek-ai/DeepSeek-R1",
-    "deepseek-ai/DeepSeek-V3": "deepseek-ai/DeepSeek-V3",
-    "deepseek-ai/deepseek-llm-67b-chat": "deepseek-ai/deepseek-llm-67b-chat",
-    "google/gemma-2-9b-it": "google/gemma-2-9b-it",
-    "google/gemma-2b-it": "google/gemma-2-27b-it",
-    "meta-llama/Llama-2-13b-chat-hf": "meta-llama/Llama-2-13b-chat-hf",
-    "meta-llama/Llama-2-7b-chat-hf": "meta-llama/Llama-2-7b-chat-hf",
-    "meta-llama/Llama-3.2-11B-Vision-Instruct": "meta-llama/Llama-Vision-Free",
-    "meta-llama/Llama-3.2-3B-Instruct": "meta-llama/Llama-3.2-3B-Instruct-Turbo",
-    "meta-llama/Llama-3.2-90B-Vision-Instruct": "meta-llama/Llama-3.2-90B-Vision-Instruct-Turbo",
-    "meta-llama/Llama-3.3-70B-Instruct": "meta-llama/Llama-3.3-70B-Instruct-Turbo",
-    "meta-llama/Meta-Llama-3-70B-Instruct": "meta-llama/Llama-3-70b-chat-hf",
-    "meta-llama/Meta-Llama-3-8B-Instruct": "meta-llama/Meta-Llama-3-8B-Instruct-Turbo",
-    "meta-llama/Meta-Llama-3.1-405B-Instruct": "meta-llama/Llama-3.2-11B-Vision-Instruct-Turbo",
-    "meta-llama/Meta-Llama-3.1-70B-Instruct": "meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo",
-    "meta-llama/Meta-Llama-3.1-8B-Instruct": "meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo-128K",
-    "microsoft/WizardLM-2-8x22B": "microsoft/WizardLM-2-8x22B",
-    "mistralai/Mistral-7B-Instruct-v0.3": "mistralai/Mistral-7B-Instruct-v0.3",
-    "mistralai/Mistral-Small-24B-Instruct-2501": "mistralai/Mistral-Small-24B-Instruct-2501",
-    "mistralai/Mixtral-8x22B-Instruct-v0.1": "mistralai/Mixtral-8x22B-Instruct-v0.1",
-    "mistralai/Mixtral-8x7B-Instruct-v0.1": "mistralai/Mixtral-8x7B-Instruct-v0.1",
-    "NousResearch/Nous-Hermes-2-Mixtral-8x7B-DPO": "NousResearch/Nous-Hermes-2-Mixtral-8x7B-DPO",
-    "nvidia/Llama-3.1-Nemotron-70B-Instruct-HF": "nvidia/Llama-3.1-Nemotron-70B-Instruct-HF",
-    "Qwen/Qwen2-72B-Instruct": "Qwen/Qwen2-72B-Instruct",
-    "Qwen/Qwen2.5-72B-Instruct": "Qwen/Qwen2.5-72B-Instruct-Turbo",
-    "Qwen/Qwen2.5-7B-Instruct": "Qwen/Qwen2.5-7B-Instruct-Turbo",
-    "Qwen/Qwen2.5-Coder-32B-Instruct": "Qwen/Qwen2.5-Coder-32B-Instruct",
-    "Qwen/QwQ-32B-Preview": "Qwen/QwQ-32B-Preview",
-    "scb10x/llama-3-typhoon-v1.5-8b-instruct": "scb10x/scb10x-llama3-typhoon-v1-5-8b-instruct",
-    "scb10x/llama-3-typhoon-v1.5x-70b-instruct-awq": "scb10x/scb10x-llama3-typhoon-v1-5x-4f316"
-  },
-  "text-generation": {
-    "meta-llama/Llama-2-70b-hf": "meta-llama/Llama-2-70b-hf",
-    "mistralai/Mixtral-8x7B-v0.1": "mistralai/Mixtral-8x7B-v0.1"
-  }
-};
+
+// src/providers/fireworks-ai.ts
+var FIREWORKS_AI_API_BASE_URL = "https://api.fireworks.ai/inference";
 
 // src/lib/isUrl.ts
 function isUrl(modelOrUrl) {
@@ -232,10 +122,70 @@ function isUrl(modelOrUrl) {
 
 // package.json
 var name = "@huggingface/inference";
-var version = "3.1.6";
+var version = "3.3.0";
+
+// src/providers/consts.ts
+var HARDCODED_MODEL_ID_MAPPING = {
+  /**
+   * "HF model ID" => "Model ID on Inference Provider's side"
+   *
+   * Example:
+   * "Qwen/Qwen2.5-Coder-32B-Instruct": "Qwen2.5-Coder-32B-Instruct",
+   */
+  "fal-ai": {},
+  "fireworks-ai": {},
+  "hf-inference": {},
+  replicate: {},
+  sambanova: {},
+  together: {}
+};
+
+// src/lib/getProviderModelId.ts
+var inferenceProviderMappingCache = /* @__PURE__ */ new Map();
+async function getProviderModelId(params, args, options = {}) {
+  if (params.provider === "hf-inference") {
+    return params.model;
+  }
+  if (!options.taskHint) {
+    throw new Error("taskHint must be specified when using a third-party provider");
+  }
+  const task = options.taskHint === "text-generation" && options.chatCompletion ? "conversational" : options.taskHint;
+  if (HARDCODED_MODEL_ID_MAPPING[params.provider]?.[params.model]) {
+    return HARDCODED_MODEL_ID_MAPPING[params.provider][params.model];
+  }
+  let inferenceProviderMapping;
+  if (inferenceProviderMappingCache.has(params.model)) {
+    inferenceProviderMapping = inferenceProviderMappingCache.get(params.model);
+  } else {
+    inferenceProviderMapping = await (options?.fetch ?? fetch)(
+      `${HF_HUB_URL}/api/models/${params.model}?expand[]=inferenceProviderMapping`,
+      {
+        headers: args.accessToken?.startsWith("hf_") ? { Authorization: `Bearer ${args.accessToken}` } : {}
+      }
+    ).then((resp) => resp.json()).then((json) => json.inferenceProviderMapping).catch(() => null);
+  }
+  if (!inferenceProviderMapping) {
+    throw new Error(`We have not been able to find inference provider information for model ${params.model}.`);
+  }
+  const providerMapping = inferenceProviderMapping[params.provider];
+  if (providerMapping) {
+    if (providerMapping.task !== task) {
+      throw new Error(
+        `Model ${params.model} is not supported for task ${task} and provider ${params.provider}. Supported task: ${providerMapping.task}.`
+      );
+    }
+    if (providerMapping.status === "staging") {
+      console.warn(
+        `Model ${params.model} is in staging mode for provider ${params.provider}. Meant for test purposes only.`
+      );
+    }
+    return providerMapping.providerId;
+  }
+  throw new Error(`Model ${params.model} is not supported provider ${params.provider}.`);
+}
 
 // src/lib/makeRequestOptions.ts
-var HF_HUB_INFERENCE_PROXY_TEMPLATE = `${HF_HUB_URL}/api/inference-proxy/{{PROVIDER}}`;
+var HF_HUB_INFERENCE_PROXY_TEMPLATE = `${HF_ROUTER_URL}/{{PROVIDER}}`;
 var tasks = null;
 async function makeRequestOptions(args, options) {
   const { accessToken, endpointUrl, provider: maybeProvider, model: maybeModel, ...remainingArgs } = args;
@@ -251,16 +201,15 @@ async function makeRequestOptions(args, options) {
   if (maybeModel && isUrl(maybeModel)) {
     throw new Error(`Model URLs are no longer supported. Use endpointUrl instead.`);
   }
-  let model;
-  if (!maybeModel) {
-    if (taskHint) {
-      model = mapModel({ model: await loadDefaultModel(taskHint), provider, taskHint, chatCompletion: chatCompletion2 });
-    } else {
-      throw new Error("No model provided, and no default model found for this task");
-    }
-  } else {
-    model = mapModel({ model: maybeModel, provider, taskHint, chatCompletion: chatCompletion2 });
+  if (!maybeModel && !taskHint) {
+    throw new Error("No model provided, and no task has been specified.");
   }
+  const hfModel = maybeModel ?? await loadDefaultModel(taskHint);
+  const model = await getProviderModelId({ model: hfModel, provider }, args, {
+    taskHint,
+    chatCompletion: chatCompletion2,
+    fetch: options?.fetch
+  });
   const authMethod = accessToken ? accessToken.startsWith("hf_") ? "hf-token" : "provider-key" : includeCredentials === "include" ? "credentials-include" : "none";
   const url = endpointUrl ? chatCompletion2 ? endpointUrl + `/v1/chat/completions` : endpointUrl : makeUrl({
     authMethod,
@@ -316,31 +265,6 @@ async function makeRequestOptions(args, options) {
   };
   return { url, info };
 }
-function mapModel(params) {
-  if (params.provider === "hf-inference") {
-    return params.model;
-  }
-  if (!params.taskHint) {
-    throw new Error("taskHint must be specified when using a third-party provider");
-  }
-  const task = params.taskHint === "text-generation" && params.chatCompletion ? "conversational" : params.taskHint;
-  const model = (() => {
-    switch (params.provider) {
-      case "fal-ai":
-        return FAL_AI_SUPPORTED_MODEL_IDS[task]?.[params.model];
-      case "replicate":
-        return REPLICATE_SUPPORTED_MODEL_IDS[task]?.[params.model];
-      case "sambanova":
-        return SAMBANOVA_SUPPORTED_MODEL_IDS[task]?.[params.model];
-      case "together":
-        return TOGETHER_SUPPORTED_MODEL_IDS[task]?.[params.model];
-    }
-  })();
-  if (!model) {
-    throw new Error(`Model ${params.model} is not supported for task ${task} and provider ${params.provider}`);
-  }
-  return model;
-}
 function makeUrl(params) {
   if (params.authMethod === "none" && params.provider !== "hf-inference") {
     throw new Error("Authentication is required when requesting a third-party provider. Please provide accessToken");
@@ -378,6 +302,13 @@ function makeUrl(params) {
       }
       return baseUrl;
     }
+    case "fireworks-ai": {
+      const baseUrl = shouldProxy ? HF_HUB_INFERENCE_PROXY_TEMPLATE.replace("{{PROVIDER}}", params.provider) : FIREWORKS_AI_API_BASE_URL;
+      if (params.taskHint === "text-generation" && params.chatCompletion) {
+        return `${baseUrl}/v1/chat/completions`;
+      }
+      return baseUrl;
+    }
     default: {
       const baseUrl = HF_HUB_INFERENCE_PROXY_TEMPLATE.replaceAll("{{PROVIDER}}", "hf-inference");
       const url = params.forceTask ? `${baseUrl}/pipeline/${params.forceTask}/${params.model}` : `${baseUrl}/models/${params.model}`;
@@ -1356,17 +1287,20 @@ var HfInferenceEndpoint = class {
 };
 
 // src/types.ts
-var INFERENCE_PROVIDERS = ["fal-ai", "replicate", "sambanova", "together", "hf-inference"];
+var INFERENCE_PROVIDERS = [
+  "fal-ai",
+  "fireworks-ai",
+  "hf-inference",
+  "replicate",
+  "sambanova",
+  "together"
+];
 // Annotate the CommonJS export names for ESM import in node:
 0 && (module.exports = {
-  FAL_AI_SUPPORTED_MODEL_IDS,
   HfInference,
   HfInferenceEndpoint,
   INFERENCE_PROVIDERS,
   InferenceOutputError,
-  REPLICATE_SUPPORTED_MODEL_IDS,
-  SAMBANOVA_SUPPORTED_MODEL_IDS,
-  TOGETHER_SUPPORTED_MODEL_IDS,
   audioClassification,
   audioToAudio,
   automaticSpeechRecognition,

package/dist/index.js

@@ -43,128 +43,22 @@ __export(tasks_exports, {
 
 // src/config.ts
 var HF_HUB_URL = "https://huggingface.co";
+var HF_ROUTER_URL = "https://router.huggingface.co";
 
 // src/providers/fal-ai.ts
 var FAL_AI_API_BASE_URL = "https://fal.run";
-var FAL_AI_SUPPORTED_MODEL_IDS = {
-  "text-to-image": {
-    "black-forest-labs/FLUX.1-schnell": "fal-ai/flux/schnell",
-    "black-forest-labs/FLUX.1-dev": "fal-ai/flux/dev",
-    "playgroundai/playground-v2.5-1024px-aesthetic": "fal-ai/playground-v25",
-    "ByteDance/SDXL-Lightning": "fal-ai/lightning-models",
-    "PixArt-alpha/PixArt-Sigma-XL-2-1024-MS": "fal-ai/pixart-sigma",
-    "stabilityai/stable-diffusion-3-medium": "fal-ai/stable-diffusion-v3-medium",
-    "Warlord-K/Sana-1024": "fal-ai/sana",
-    "fal/AuraFlow-v0.2": "fal-ai/aura-flow",
-    "stabilityai/stable-diffusion-3.5-large": "fal-ai/stable-diffusion-v35-large",
-    "stabilityai/stable-diffusion-3.5-large-turbo": "fal-ai/stable-diffusion-v35-large/turbo",
-    "stabilityai/stable-diffusion-3.5-medium": "fal-ai/stable-diffusion-v35-medium",
-    "Kwai-Kolors/Kolors": "fal-ai/kolors"
-  },
-  "automatic-speech-recognition": {
-    "openai/whisper-large-v3": "fal-ai/whisper"
-  },
-  "text-to-video": {
-    "genmo/mochi-1-preview": "fal-ai/mochi-v1",
-    "tencent/HunyuanVideo": "fal-ai/hunyuan-video",
-    "THUDM/CogVideoX-5b": "fal-ai/cogvideox-5b",
-    "Lightricks/LTX-Video": "fal-ai/ltx-video"
-  }
-};
 
 // src/providers/replicate.ts
 var REPLICATE_API_BASE_URL = "https://api.replicate.com";
-var REPLICATE_SUPPORTED_MODEL_IDS = {
-  "text-to-image": {
-    "black-forest-labs/FLUX.1-dev": "black-forest-labs/flux-dev",
-    "black-forest-labs/FLUX.1-schnell": "black-forest-labs/flux-schnell",
-    "ByteDance/Hyper-SD": "bytedance/hyper-flux-16step:382cf8959fb0f0d665b26e7e80b8d6dc3faaef1510f14ce017e8c732bb3d1eb7",
-    "ByteDance/SDXL-Lightning": "bytedance/sdxl-lightning-4step:5599ed30703defd1d160a25a63321b4dec97101d98b4674bcc56e41f62f35637",
-    "playgroundai/playground-v2.5-1024px-aesthetic": "playgroundai/playground-v2.5-1024px-aesthetic:a45f82a1382bed5c7aeb861dac7c7d191b0fdf74d8d57c4a0e6ed7d4d0bf7d24",
-    "stabilityai/stable-diffusion-3.5-large-turbo": "stability-ai/stable-diffusion-3.5-large-turbo",
-    "stabilityai/stable-diffusion-3.5-large": "stability-ai/stable-diffusion-3.5-large",
-    "stabilityai/stable-diffusion-3.5-medium": "stability-ai/stable-diffusion-3.5-medium",
-    "stabilityai/stable-diffusion-xl-base-1.0": "stability-ai/sdxl:7762fd07cf82c948538e41f63f77d685e02b063e37e496e96eefd46c929f9bdc"
-  },
-  "text-to-speech": {
-    "OuteAI/OuteTTS-0.3-500M": "jbilcke/oute-tts:3c645149db020c85d080e2f8cfe482a0e68189a922cde964fa9e80fb179191f3",
-    "hexgrad/Kokoro-82M": "jaaari/kokoro-82m:dfdf537ba482b029e0a761699e6f55e9162cfd159270bfe0e44857caa5f275a6"
-  },
-  "text-to-video": {
-    "genmo/mochi-1-preview": "genmoai/mochi-1:1944af04d098ef69bed7f9d335d102e652203f268ec4aaa2d836f6217217e460"
-  }
-};
 
 // src/providers/sambanova.ts
 var SAMBANOVA_API_BASE_URL = "https://api.sambanova.ai";
-var SAMBANOVA_SUPPORTED_MODEL_IDS = {
-  /** Chat completion / conversational */
-  conversational: {
-    "deepseek-ai/DeepSeek-Distill-R1-Llama-70B": "DeepSeek-Distill-R1-Llama-70B",
-    "Qwen/Qwen2.5-Coder-32B-Instruct": "Qwen2.5-Coder-32B-Instruct",
-    "Qwen/Qwen2.5-72B-Instruct": "Qwen2.5-72B-Instruct",
-    "Qwen/QwQ-32B-Preview": "QwQ-32B-Preview",
-    "meta-llama/Llama-3.3-70B-Instruct": "Meta-Llama-3.3-70B-Instruct",
-    "meta-llama/Llama-3.2-1B-Instruct": "Meta-Llama-3.2-1B-Instruct",
-    "meta-llama/Llama-3.2-3B-Instruct": "Meta-Llama-3.2-3B-Instruct",
-    "meta-llama/Llama-3.2-11B-Vision-Instruct": "Llama-3.2-11B-Vision-Instruct",
-    "meta-llama/Llama-3.2-90B-Vision-Instruct": "Llama-3.2-90B-Vision-Instruct",
-    "meta-llama/Llama-3.1-8B-Instruct": "Meta-Llama-3.1-8B-Instruct",
-    "meta-llama/Llama-3.1-70B-Instruct": "Meta-Llama-3.1-70B-Instruct",
-    "meta-llama/Llama-3.1-405B-Instruct": "Meta-Llama-3.1-405B-Instruct",
-    "meta-llama/Llama-Guard-3-8B": "Meta-Llama-Guard-3-8B"
-  }
-};
 
 // src/providers/together.ts
 var TOGETHER_API_BASE_URL = "https://api.together.xyz";
-var TOGETHER_SUPPORTED_MODEL_IDS = {
-  "text-to-image": {
-    "black-forest-labs/FLUX.1-Canny-dev": "black-forest-labs/FLUX.1-canny",
-    "black-forest-labs/FLUX.1-Depth-dev": "black-forest-labs/FLUX.1-depth",
-    "black-forest-labs/FLUX.1-dev": "black-forest-labs/FLUX.1-dev",
-    "black-forest-labs/FLUX.1-Redux-dev": "black-forest-labs/FLUX.1-redux",
-    "black-forest-labs/FLUX.1-schnell": "black-forest-labs/FLUX.1-pro",
-    "stabilityai/stable-diffusion-xl-base-1.0": "stabilityai/stable-diffusion-xl-base-1.0"
-  },
-  conversational: {
-    "databricks/dbrx-instruct": "databricks/dbrx-instruct",
-    "deepseek-ai/DeepSeek-R1": "deepseek-ai/DeepSeek-R1",
-    "deepseek-ai/DeepSeek-V3": "deepseek-ai/DeepSeek-V3",
-    "deepseek-ai/deepseek-llm-67b-chat": "deepseek-ai/deepseek-llm-67b-chat",
-    "google/gemma-2-9b-it": "google/gemma-2-9b-it",
-    "google/gemma-2b-it": "google/gemma-2-27b-it",
-    "meta-llama/Llama-2-13b-chat-hf": "meta-llama/Llama-2-13b-chat-hf",
-    "meta-llama/Llama-2-7b-chat-hf": "meta-llama/Llama-2-7b-chat-hf",
-    "meta-llama/Llama-3.2-11B-Vision-Instruct": "meta-llama/Llama-Vision-Free",
-    "meta-llama/Llama-3.2-3B-Instruct": "meta-llama/Llama-3.2-3B-Instruct-Turbo",
-    "meta-llama/Llama-3.2-90B-Vision-Instruct": "meta-llama/Llama-3.2-90B-Vision-Instruct-Turbo",
-    "meta-llama/Llama-3.3-70B-Instruct": "meta-llama/Llama-3.3-70B-Instruct-Turbo",
-    "meta-llama/Meta-Llama-3-70B-Instruct": "meta-llama/Llama-3-70b-chat-hf",
-    "meta-llama/Meta-Llama-3-8B-Instruct": "meta-llama/Meta-Llama-3-8B-Instruct-Turbo",
-    "meta-llama/Meta-Llama-3.1-405B-Instruct": "meta-llama/Llama-3.2-11B-Vision-Instruct-Turbo",
-    "meta-llama/Meta-Llama-3.1-70B-Instruct": "meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo",
-    "meta-llama/Meta-Llama-3.1-8B-Instruct": "meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo-128K",
-    "microsoft/WizardLM-2-8x22B": "microsoft/WizardLM-2-8x22B",
-    "mistralai/Mistral-7B-Instruct-v0.3": "mistralai/Mistral-7B-Instruct-v0.3",
-    "mistralai/Mistral-Small-24B-Instruct-2501": "mistralai/Mistral-Small-24B-Instruct-2501",
-    "mistralai/Mixtral-8x22B-Instruct-v0.1": "mistralai/Mixtral-8x22B-Instruct-v0.1",
-    "mistralai/Mixtral-8x7B-Instruct-v0.1": "mistralai/Mixtral-8x7B-Instruct-v0.1",
-    "NousResearch/Nous-Hermes-2-Mixtral-8x7B-DPO": "NousResearch/Nous-Hermes-2-Mixtral-8x7B-DPO",
-    "nvidia/Llama-3.1-Nemotron-70B-Instruct-HF": "nvidia/Llama-3.1-Nemotron-70B-Instruct-HF",
-    "Qwen/Qwen2-72B-Instruct": "Qwen/Qwen2-72B-Instruct",
-    "Qwen/Qwen2.5-72B-Instruct": "Qwen/Qwen2.5-72B-Instruct-Turbo",
-    "Qwen/Qwen2.5-7B-Instruct": "Qwen/Qwen2.5-7B-Instruct-Turbo",
-    "Qwen/Qwen2.5-Coder-32B-Instruct": "Qwen/Qwen2.5-Coder-32B-Instruct",
-    "Qwen/QwQ-32B-Preview": "Qwen/QwQ-32B-Preview",
-    "scb10x/llama-3-typhoon-v1.5-8b-instruct": "scb10x/scb10x-llama3-typhoon-v1-5-8b-instruct",
-    "scb10x/llama-3-typhoon-v1.5x-70b-instruct-awq": "scb10x/scb10x-llama3-typhoon-v1-5x-4f316"
-  },
-  "text-generation": {
-    "meta-llama/Llama-2-70b-hf": "meta-llama/Llama-2-70b-hf",
-    "mistralai/Mixtral-8x7B-v0.1": "mistralai/Mixtral-8x7B-v0.1"
-  }
-};
+
+// src/providers/fireworks-ai.ts
+var FIREWORKS_AI_API_BASE_URL = "https://api.fireworks.ai/inference";
 
 // src/lib/isUrl.ts
 function isUrl(modelOrUrl) {
@@ -173,10 +67,70 @@ function isUrl(modelOrUrl) {
 
 // package.json
 var name = "@huggingface/inference";
-var version = "3.1.6";
+var version = "3.3.0";
+
+// src/providers/consts.ts
+var HARDCODED_MODEL_ID_MAPPING = {
+  /**
+   * "HF model ID" => "Model ID on Inference Provider's side"
+   *
+   * Example:
+   * "Qwen/Qwen2.5-Coder-32B-Instruct": "Qwen2.5-Coder-32B-Instruct",
+   */
+  "fal-ai": {},
+  "fireworks-ai": {},
+  "hf-inference": {},
+  replicate: {},
+  sambanova: {},
+  together: {}
+};
+
+// src/lib/getProviderModelId.ts
+var inferenceProviderMappingCache = /* @__PURE__ */ new Map();
+async function getProviderModelId(params, args, options = {}) {
+  if (params.provider === "hf-inference") {
+    return params.model;
+  }
+  if (!options.taskHint) {
+    throw new Error("taskHint must be specified when using a third-party provider");
+  }
+  const task = options.taskHint === "text-generation" && options.chatCompletion ? "conversational" : options.taskHint;
+  if (HARDCODED_MODEL_ID_MAPPING[params.provider]?.[params.model]) {
+    return HARDCODED_MODEL_ID_MAPPING[params.provider][params.model];
+  }
+  let inferenceProviderMapping;
+  if (inferenceProviderMappingCache.has(params.model)) {
+    inferenceProviderMapping = inferenceProviderMappingCache.get(params.model);
+  } else {
+    inferenceProviderMapping = await (options?.fetch ?? fetch)(
+      `${HF_HUB_URL}/api/models/${params.model}?expand[]=inferenceProviderMapping`,
+      {
+        headers: args.accessToken?.startsWith("hf_") ? { Authorization: `Bearer ${args.accessToken}` } : {}
+      }
+    ).then((resp) => resp.json()).then((json) => json.inferenceProviderMapping).catch(() => null);
+  }
+  if (!inferenceProviderMapping) {
+    throw new Error(`We have not been able to find inference provider information for model ${params.model}.`);
+  }
+  const providerMapping = inferenceProviderMapping[params.provider];
+  if (providerMapping) {
+    if (providerMapping.task !== task) {
+      throw new Error(
+        `Model ${params.model} is not supported for task ${task} and provider ${params.provider}. Supported task: ${providerMapping.task}.`
+      );
+    }
+    if (providerMapping.status === "staging") {
+      console.warn(
+        `Model ${params.model} is in staging mode for provider ${params.provider}. Meant for test purposes only.`
+      );
+    }
+    return providerMapping.providerId;
+  }
+  throw new Error(`Model ${params.model} is not supported provider ${params.provider}.`);
+}
 
 // src/lib/makeRequestOptions.ts
-var HF_HUB_INFERENCE_PROXY_TEMPLATE = `${HF_HUB_URL}/api/inference-proxy/{{PROVIDER}}`;
+var HF_HUB_INFERENCE_PROXY_TEMPLATE = `${HF_ROUTER_URL}/{{PROVIDER}}`;
 var tasks = null;
 async function makeRequestOptions(args, options) {
   const { accessToken, endpointUrl, provider: maybeProvider, model: maybeModel, ...remainingArgs } = args;
@@ -192,16 +146,15 @@ async function makeRequestOptions(args, options) {
   if (maybeModel && isUrl(maybeModel)) {
     throw new Error(`Model URLs are no longer supported. Use endpointUrl instead.`);
   }
-  let model;
-  if (!maybeModel) {
-    if (taskHint) {
-      model = mapModel({ model: await loadDefaultModel(taskHint), provider, taskHint, chatCompletion: chatCompletion2 });
-    } else {
-      throw new Error("No model provided, and no default model found for this task");
-    }
-  } else {
-    model = mapModel({ model: maybeModel, provider, taskHint, chatCompletion: chatCompletion2 });
+  if (!maybeModel && !taskHint) {
+    throw new Error("No model provided, and no task has been specified.");
   }
+  const hfModel = maybeModel ?? await loadDefaultModel(taskHint);
+  const model = await getProviderModelId({ model: hfModel, provider }, args, {
+    taskHint,
+    chatCompletion: chatCompletion2,
+    fetch: options?.fetch
+  });
   const authMethod = accessToken ? accessToken.startsWith("hf_") ? "hf-token" : "provider-key" : includeCredentials === "include" ? "credentials-include" : "none";
   const url = endpointUrl ? chatCompletion2 ? endpointUrl + `/v1/chat/completions` : endpointUrl : makeUrl({
     authMethod,
@@ -257,31 +210,6 @@ async function makeRequestOptions(args, options) {
   };
   return { url, info };
 }
-function mapModel(params) {
-  if (params.provider === "hf-inference") {
-    return params.model;
-  }
-  if (!params.taskHint) {
-    throw new Error("taskHint must be specified when using a third-party provider");
-  }
-  const task = params.taskHint === "text-generation" && params.chatCompletion ? "conversational" : params.taskHint;
-  const model = (() => {
-    switch (params.provider) {
-      case "fal-ai":
-        return FAL_AI_SUPPORTED_MODEL_IDS[task]?.[params.model];
-      case "replicate":
-        return REPLICATE_SUPPORTED_MODEL_IDS[task]?.[params.model];
-      case "sambanova":
-        return SAMBANOVA_SUPPORTED_MODEL_IDS[task]?.[params.model];
-      case "together":
-        return TOGETHER_SUPPORTED_MODEL_IDS[task]?.[params.model];
-    }
-  })();
-  if (!model) {
-    throw new Error(`Model ${params.model} is not supported for task ${task} and provider ${params.provider}`);
-  }
-  return model;
-}
 function makeUrl(params) {
   if (params.authMethod === "none" && params.provider !== "hf-inference") {
     throw new Error("Authentication is required when requesting a third-party provider. Please provide accessToken");
@@ -319,6 +247,13 @@ function makeUrl(params) {
       }
       return baseUrl;
     }
+    case "fireworks-ai": {
+      const baseUrl = shouldProxy ? HF_HUB_INFERENCE_PROXY_TEMPLATE.replace("{{PROVIDER}}", params.provider) : FIREWORKS_AI_API_BASE_URL;
+      if (params.taskHint === "text-generation" && params.chatCompletion) {
+        return `${baseUrl}/v1/chat/completions`;
+      }
+      return baseUrl;
+    }
     default: {
       const baseUrl = HF_HUB_INFERENCE_PROXY_TEMPLATE.replaceAll("{{PROVIDER}}", "hf-inference");
       const url = params.forceTask ? `${baseUrl}/pipeline/${params.forceTask}/${params.model}` : `${baseUrl}/models/${params.model}`;
@@ -1297,16 +1232,19 @@ var HfInferenceEndpoint = class {
 };
 
 // src/types.ts
-var INFERENCE_PROVIDERS = ["fal-ai", "replicate", "sambanova", "together", "hf-inference"];
+var INFERENCE_PROVIDERS = [
+  "fal-ai",
+  "fireworks-ai",
+  "hf-inference",
+  "replicate",
+  "sambanova",
+  "together"
+];
 export {
-  FAL_AI_SUPPORTED_MODEL_IDS,
   HfInference,
   HfInferenceEndpoint,
   INFERENCE_PROVIDERS,
   InferenceOutputError,
-  REPLICATE_SUPPORTED_MODEL_IDS,
-  SAMBANOVA_SUPPORTED_MODEL_IDS,
-  TOGETHER_SUPPORTED_MODEL_IDS,
   audioClassification,
   audioToAudio,
   automaticSpeechRecognition,