Merge remote-tracking branch 'hf/main' into ovhcloud-inference-provider

Author: Fabien Ric

.github/workflows/mcp-client-publish.yml

@@ -0,0 +1,72 @@
+name: MCP Client - Version and Release
+
+on:
+  workflow_dispatch:
+    inputs:
+      newversion:
+        type: choice
+        description: "Semantic Version Bump Type"
+        default: patch
+        options:
+          - patch
+          - minor
+          - major
+
+concurrency:
+  group: "push-to-main"
+
+defaults:
+  run:
+    working-directory: packages/mcp-client
+
+jobs:
+  version_and_release:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v3
+        with:
+          token: ${{ secrets.BOT_ACCESS_TOKEN }}
+      - run: npm install -g corepack@latest && corepack enable
+      - uses: actions/setup-node@v3
+        with:
+          node-version: "20"
+          cache: "pnpm"
+          cache-dependency-path: |
+            packages/mcp-client/pnpm-lock.yaml
+            packages/doc-internal/pnpm-lock.yaml
+          registry-url: "https://registry.npmjs.org"
+      - run: pnpm install
+      - run: git config --global user.name machineuser
+      - run: git config --global user.email [email protected]
+      - run: |
+          PACKAGE_VERSION=$(node -p "require('./package.json').version")
+          BUMPED_VERSION=$(node -p "require('semver').inc('$PACKAGE_VERSION', '${{ github.event.inputs.newversion }}')")
+          # Update package.json with the new version
+          node -e "const fs = require('fs'); const package = JSON.parse(fs.readFileSync('./package.json')); package.version = '$BUMPED_VERSION'; fs.writeFileSync('./package.json', JSON.stringify(package, null, '\t') + '\n');"
+          pnpm --filter doc-internal run fix-cdn-versions
+          git add ../..
+          git commit -m "🔖 @huggingface/mcp-client $BUMPED_VERSION"
+          git tag "mcp-client-v$BUMPED_VERSION"
+
+      # Add checks for dependencies if needed, similar to hub-publish.yml
+      - name: "Check Deps are published before publishing this package"
+        run: pnpm -w check-deps inference && pnpm -w check-deps tasks
+
+      - run: pnpm publish --no-git-checks .
+        env:
+          NODE_AUTH_TOKEN: ${{ secrets.NPM_TOKEN }}
+      - run: git pull --rebase && git push --follow-tags
+      # hack - reuse actions/setup-node@v3 just to set a new registry
+      - uses: actions/setup-node@v3
+        with:
+          node-version: "20"
+          registry-url: "https://npm.pkg.github.com"
+      # Disable for now, until github supports PATs for writing github packages (https://github.com/github/roadmap/issues/558)
+      # - run: pnpm publish --no-git-checks .
+      #   env:
+      #     NODE_AUTH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+      - name: "Update Doc"
+        uses: peter-evans/repository-dispatch@v2
+        with:
+          event-type: doc-build
+          token: ${{ secrets.BOT_ACCESS_TOKEN }}

.github/workflows/test.yml

@@ -38,7 +38,7 @@ jobs:
           pnpm --filter ...[${{ steps.since.outputs.SINCE }}]... build
 
       - name: Test
-        run: VCR_MODE=playback pnpm --filter ...[${{ steps.since.outputs.SINCE }}] test
+        run: pnpm --filter ...[${{ steps.since.outputs.SINCE }}] test
         env:
           HF_TOKEN: ${{ secrets.HF_TOKEN }}
 
@@ -76,7 +76,7 @@ jobs:
           pnpm --filter ...[${{ steps.since.outputs.SINCE }}]... build
 
       - name: Test in browser
-        run: VCR_MODE=playback pnpm --filter ...[${{ steps.since.outputs.SINCE }}] test:browser
+        run: pnpm --filter ...[${{ steps.since.outputs.SINCE }}] test:browser
         env:
           HF_TOKEN: ${{ secrets.HF_TOKEN }}
 

README.md

@@ -97,7 +97,7 @@ You can run our packages with vanilla JS, without any bundler, by using a CDN or
 
 ```html
 <script type="module">
-    import { InferenceClient } from 'https://cdn.jsdelivr.net/npm/@huggingface/inference@3.8.1/+esm';
+    import { InferenceClient } from 'https://cdn.jsdelivr.net/npm/@huggingface/inference@3.9.2/+esm';
     import { createRepo, commit, deleteRepo, listFiles } from "https://cdn.jsdelivr.net/npm/@huggingface/[email protected]/+esm";
 </script>
 ```

docs/_toctree.yml

@@ -14,6 +14,11 @@
       title: Interact with the Hub
     - local: hub/modules
       title: API Reference
+- title: "@huggingface/mcp-client"
+  isExpanded: true
+  sections:
+    - local: mcp-client/README
+      title: Simple MCP Client and smol Agent built on top of Inference Client
 - title: "@huggingface/agent"
   isExpanded: true
   sections:

package.json

@@ -11,9 +11,9 @@
 		"check-deps": "tsx scripts/check-deps.ts"
 	},
 	"devDependencies": {
+		"@types/node": "^22.14.1",
 		"@typescript-eslint/eslint-plugin": "^7.2.0",
 		"@typescript-eslint/parser": "^7.2.0",
-		"@types/node": "^18.16.1",
 		"@vitest/browser": "^0.34.6",
 		"eslint": "^8.57.0",
 		"eslint-config-prettier": "^9.0.0",

packages/doc-internal/package.json

@@ -5,7 +5,7 @@
 	"description": "Package to generate doc for other @huggingface packages",
 	"private": true,
 	"scripts": {
-		"start": "pnpm run fix-cdn-versions && pnpm run doc-hub && pnpm run doc-inference && pnpm run doc-agents && pnpm run doc-space-header && pnpm run doc-gguf && cp ../../README.md ../../docs/index.md && pnpm run update-toc && pnpm run fix-md-links && pnpm run fix-md-headinghashlinks",
+		"start": "pnpm run fix-cdn-versions && pnpm run doc-hub && pnpm run doc-inference && pnpm run doc-agents && pnpm run doc-space-header && pnpm run doc-gguf && pnpm run doc-mcp-client && cp ../../README.md ../../docs/index.md && pnpm run update-toc && pnpm run fix-md-links && pnpm run fix-md-headinghashlinks",
 		"lint": "eslint --quiet --fix --ext .cjs,.ts .",
 		"lint:check": "eslint --ext .cjs,.ts .",
 		"format": "prettier --write .",
@@ -14,6 +14,7 @@
 		"doc-inference": "typedoc --tsconfig ../inference/tsconfig.json --githubPages false --plugin typedoc-plugin-markdown --out ../../docs/inference --hideBreadcrumbs --hideInPageTOC --sourceLinkTemplate https://github.com/huggingface/huggingface.js/blob/main/{path}#L{line} ../inference/src/index.ts",
 		"doc-agents": "typedoc --tsconfig ../agents/tsconfig.json --githubPages false --plugin typedoc-plugin-markdown --out ../../docs/agents --hideBreadcrumbs --hideInPageTOC --sourceLinkTemplate https://github.com/huggingface/huggingface.js/blob/main/{path}#L{line} ../agents/src/index.ts",
 		"doc-gguf": "mkdir -p ../../docs/gguf && cp ../../packages/gguf/README.md ../../docs/gguf/README.md",
+		"doc-mcp-client": "mkdir -p ../../docs/mcp-client && cp ../../packages/mcp-client/README.md ../../docs/mcp-client/README.md",
 		"doc-space-header": "mkdir -p ../../docs/space-header && cp ../../packages/space-header/README.md ../../docs/space-header/README.md",
 		"update-toc": "tsx update-toc.ts",
 		"fix-cdn-versions": "tsx fix-cdn-versions.ts",

packages/inference/README.md

@@ -48,17 +48,20 @@ You can send inference requests to third-party providers with the inference clie
 
 Currently, we support the following providers:
 - [Fal.ai](https://fal.ai)
+- [Featherless AI](https://featherless.ai)
 - [Fireworks AI](https://fireworks.ai)
 - [Hyperbolic](https://hyperbolic.xyz)
 - [Nebius](https://studio.nebius.ai)
 - [Novita](https://novita.ai/?utm_source=github_huggingface&utm_medium=github_readme&utm_campaign=link)
+- [Nscale](https://nscale.com)
+- [OVHcloud](https://endpoints.ai.cloud.ovh.net/)
 - [Replicate](https://replicate.com)
 - [Sambanova](https://sambanova.ai)
 - [Together](https://together.xyz)
 - [Blackforestlabs](https://blackforestlabs.ai)
 - [Cohere](https://cohere.com)
 - [Cerebras](https://cerebras.ai/)
-- [OVHcloud](https://endpoints.ai.cloud.ovh.net/)
+- [Groq](https://groq.com)
 
 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
@@ -77,15 +80,18 @@ When authenticated with a third-party provider key, the request is made directly
 
 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](https://huggingface.co/api/partners/fal-ai/models)
+- [Featherless AI supported models](https://huggingface.co/api/partners/featherless-ai/models)
 - [Fireworks AI supported models](https://huggingface.co/api/partners/fireworks-ai/models)
 - [Hyperbolic supported models](https://huggingface.co/api/partners/hyperbolic/models)
 - [Nebius supported models](https://huggingface.co/api/partners/nebius/models)
+- [Nscale supported models](https://huggingface.co/api/partners/nscale/models)
+- [OVHcloud supported models](https://huggingface.co/api/partners/ovhcloud/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)
 - [Cohere supported models](https://huggingface.co/api/partners/cohere/models)
 - [Cerebras supported models](https://huggingface.co/api/partners/cerebras/models)
-- [OVHcloud supported models](https://huggingface.co/api/partners/ovhcloud/models)
+- [Groq supported models](https://console.groq.com/docs/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.

packages/inference/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "@huggingface/inference",
-	"version": "3.8.1",
+	"version": "3.9.2",
 	"packageManager": "[email protected]",
 	"license": "MIT",
 	"author": "Hugging Face and Tim Mikeladze <[email protected]>",

packages/inference/src/lib/getInferenceProviderMapping.ts

@@ -8,7 +8,7 @@ import { typedInclude } from "../utils/typedInclude";
 export const inferenceProviderMappingCache = new Map<ModelId, InferenceProviderMapping>();
 
 export type InferenceProviderMapping = Partial<
-	Record<InferenceProvider, Omit<InferenceProviderModelMapping, "hfModelId" | "adapterWeightsPath">>
+	Record<InferenceProvider, Omit<InferenceProviderModelMapping, "hfModelId">>
 >;
 
 export interface InferenceProviderModelMapping {
@@ -74,22 +74,6 @@ export async function getInferenceProviderMapping(
 				`Model ${params.modelId} is in staging mode for provider ${params.provider}. Meant for test purposes only.`
 			);
 		}
-		if (providerMapping.adapter === "lora") {
-			const treeResp = await (options?.fetch ?? fetch)(`${HF_HUB_URL}/api/models/${params.modelId}/tree/main`);
-			if (!treeResp.ok) {
-				throw new Error(`Unable to fetch the model tree for ${params.modelId}.`);
-			}
-			const tree: Array<{ type: "file" | "directory"; path: string }> = await treeResp.json();
-			const adapterWeightsPath = tree.find(({ type, path }) => type === "file" && path.endsWith(".safetensors"))?.path;
-			if (!adapterWeightsPath) {
-				throw new Error(`No .safetensors file found in the model tree for ${params.modelId}.`);
-			}
-			return {
-				...providerMapping,
-				hfModelId: params.modelId,
-				adapterWeightsPath,
-			};
-		}
 		return { ...providerMapping, hfModelId: params.modelId };
 	}
 	return null;

packages/inference/src/lib/getProviderHelper.ts

@@ -2,12 +2,14 @@ import * as BlackForestLabs from "../providers/black-forest-labs";
 import * as Cerebras from "../providers/cerebras";
 import * as Cohere from "../providers/cohere";
 import * as FalAI from "../providers/fal-ai";
+import * as FeatherlessAI from "../providers/featherless-ai";
 import * as Fireworks from "../providers/fireworks-ai";
+import * as Groq from "../providers/groq";
 import * as HFInference from "../providers/hf-inference";
-
 import * as Hyperbolic from "../providers/hyperbolic";
 import * as Nebius from "../providers/nebius";
 import * as Novita from "../providers/novita";
+import * as Nscale from "../providers/nscale";
 import * as OpenAI from "../providers/openai";
 import * as OvhCloud from "../providers/ovhcloud";
 import type {
@@ -63,6 +65,10 @@ export const PROVIDERS: Record<InferenceProvider, Partial<Record<InferenceTask,
 		"text-to-video": new FalAI.FalAITextToVideoTask(),
 		"automatic-speech-recognition": new FalAI.FalAIAutomaticSpeechRecognitionTask(),
 	},
+	"featherless-ai": {
+		conversational: new FeatherlessAI.FeatherlessAIConversationalTask(),
+		"text-generation": new FeatherlessAI.FeatherlessAITextGenerationTask(),
+	},
 	"hf-inference": {
 		"text-to-image": new HFInference.HFInferenceTextToImageTask(),
 		conversational: new HFInference.HFInferenceConversationalTask(),
@@ -96,6 +102,10 @@ export const PROVIDERS: Record<InferenceProvider, Partial<Record<InferenceTask,
 	"fireworks-ai": {
 		conversational: new Fireworks.FireworksConversationalTask(),
 	},
+	groq: {
+		conversational: new Groq.GroqConversationalTask(),
+		"text-generation": new Groq.GroqTextGenerationTask(),
+	},
 	hyperbolic: {
 		"text-to-image": new Hyperbolic.HyperbolicTextToImageTask(),
 		conversational: new Hyperbolic.HyperbolicConversationalTask(),
@@ -110,6 +120,10 @@ export const PROVIDERS: Record<InferenceProvider, Partial<Record<InferenceTask,
 		conversational: new Novita.NovitaConversationalTask(),
 		"text-generation": new Novita.NovitaTextGenerationTask(),
 	},
+	nscale: {
+		"text-to-image": new Nscale.NscaleTextToImageTask(),
+		conversational: new Nscale.NscaleConversationalTask(),
+	},
 	openai: {
 		conversational: new OpenAI.OpenAIConversationalTask(),
 	},