Add fal.ai inference endpoints (#1104)

Adds fal.ai inference endpoints

---------

Co-authored-by: Aykut Kardaş <[email protected]>
Co-authored-by: SBrandeis <[email protected]>

Author: 3 people

packages/inference/src/providers/fal-ai.ts

@@ -8,6 +8,14 @@ export const FAL_AI_SUPPORTED_MODEL_IDS: ProviderMapping<FalAiId> = {
 	"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",
+		"Kwai-Kolors/Kolors": "fal-ai/kolors",
 	},
 	"automatic-speech-recognition": {
 		"openai/whisper-large-v3": "fal-ai/whisper",

packages/inference/test/HfInference.spec.ts

@@ -2,7 +2,7 @@ import { expect, it, describe, assert } from "vitest";
 
 import type { ChatCompletionStreamOutput } from "@huggingface/tasks";
 
-import { chatCompletion, HfInference } from "../src";
+import { chatCompletion, FAL_AI_SUPPORTED_MODEL_IDS, HfInference } from "../src";
 import "./vcr";
 import { readTestFile } from "./test-files";
 
@@ -775,25 +775,30 @@ describe.concurrent("HfInference", () => {
 		() => {
 			const client = new HfInference(env.HF_FAL_KEY);
 
-			it("textToImage", async () => {
-				const res = await client.textToImage({
-					model: "black-forest-labs/FLUX.1-schnell",
-					provider: "fal-ai",
-					inputs: "black forest gateau cake spelling out the words FLUX SCHNELL, tasty, food photography, dynamic shot",
+			for (const model of Object.keys(FAL_AI_SUPPORTED_MODEL_IDS["text-to-image"] ?? {})) {
+				it(`textToImage - ${model}`, async () => {
+					const res = await client.textToImage({
+						model,
+						provider: "fal-ai",
+						inputs:
+							"Extreme close-up of a single tiger eye, direct frontal view. Detailed iris and pupil. Sharp focus on eye texture and color. Natural lighting to capture authentic eye shine and depth.",
+					});
+					expect(res).toBeInstanceOf(Blob);
 				});
-				expect(res).toBeInstanceOf(Blob);
-			});
+			}
 
-			it("speechToText", async () => {
-				const res = await client.automaticSpeechRecognition({
-					model: "openai/whisper-large-v3",
-					provider: "fal-ai",
-					data: new Blob([readTestFile("sample2.wav")], { type: "audio/x-wav" }),
-				});
-				expect(res).toMatchObject({
-					text: " he has grave doubts whether sir frederick leighton's work is really greek after all and can discover in it but little of rocky ithaca",
+			for (const model of Object.keys(FAL_AI_SUPPORTED_MODEL_IDS["automatic-speech-recognition"] ?? {})) {
+				it(`automaticSpeechRecognition - ${model}`, async () => {
+					const res = await client.automaticSpeechRecognition({
+						model: model,
+						provider: "fal-ai",
+						data: new Blob([readTestFile("sample2.wav")], { type: "audio/x-wav" }),
+					});
+					expect(res).toMatchObject({
+						text: " he has grave doubts whether sir frederick leighton's work is really greek after all and can discover in it but little of rocky ithaca",
+					});
 				});
-			});
+			}
 		},
 		TIMEOUT
 	);