New `InferenceClient` endpoint type (#1813)

* work on hf inference endpoint

* working tools

* fix: debug tool

* feat: working tools with new endpoint type

* feat: move config to HfInference

* wip

* fix: hf inference endpoint

* refactor: rename endpoint

* fix: rendering

* feat: working tools on every supported model

* fix: multimodal

* refactor: rename endpoint type

* refactor: base URL

* fix: prevent simultaneous provider and baseURL usage

* fix: disable tools on qwen 2.5 vl

* feat: make sure json is correctly formatted

chart/env/prod.yaml

@@ -60,6 +60,7 @@ envVars:
           "max_new_tokens": 1024,
           "truncate": 7167
         },
+        "endpoints": [{"type" : "inference-client"}],
         "promptExamples": [
           {
             "title": "Write an email",
@@ -91,6 +92,7 @@ envVars:
           "stop": ["<|endoftext|>", "<|im_end|>"],
           "temperature": 0.6,
         },
+        "tools": true,
         "promptExamples": [
           {
             "title": "Write an email",
@@ -107,7 +109,7 @@ envVars:
         ],
         "endpoints": [
           {
-            "type": "openai",
+            "type": "inference-client",
             "baseURL": "https://api-inference.endpoints.huggingface.tech/models/Qwen/Qwen3-235B-A22B/v1"
           }
         ]
@@ -126,6 +128,7 @@ envVars:
           "max_new_tokens": 3072
         },
         "tools": true,
+        "endpoints": [{"type" : "inference-client"}],
         "promptExamples": [
           {
             "title": "Write an email",
@@ -148,12 +151,14 @@ envVars:
         "websiteUrl": "https://docs.cohere.com/docs/command-r-plus",
         "logoUrl": "https://huggingface.co/datasets/huggingchat/models-logo/resolve/main/cohere-logo.png",
         "tools": true,
+        "systemRoleSupported": false,
         "parameters": {
           "stop": ["<|END_OF_TURN_TOKEN|>", "<|im_end|>"],
           "truncate": 28672,
           "max_new_tokens": 2048,
           "temperature": 0.3
         },
+        "endpoints": [{"type" : "inference-client"}],
         "promptExamples": [
           {
             "title": "Generate image",
@@ -180,6 +185,7 @@ envVars:
           "beginToken": "",
           "endToken": "</think>"
         },
+        "tools": true,
         "promptExamples": [
           {
             "title": "Rs in strawberry",
@@ -196,8 +202,8 @@ envVars:
         ],
         "endpoints": [
           {
-            "type": "openai",
-            "baseURL": "https://api-inference.endpoints.huggingface.tech/models/deepseek-ai/DeepSeek-R1-Distill-Qwen-32B/v1"
+            "type": "inference-client",
+            "baseURL": "https://internal.api-inference.huggingface.co/models/deepseek-ai/DeepSeek-R1-Distill-Qwen-32B/v1"
           }
         ]
       },
@@ -229,13 +235,14 @@ envVars:
         ],
         "endpoints": [
           {
-            "type": "openai",
+            "type": "inference-client",
             "baseURL": "https://api-inference.endpoints.huggingface.tech/models/nvidia/Llama-3.1-Nemotron-70B-Instruct-HF/v1"
           }
         ]
       },
       {
         "name": "Qwen/QwQ-32B",
+        "tools": true,
         "preprompt": "You are a helpful and harmless assistant. You are Qwen developed by Alibaba. You should think step-by-step.",
         "modelUrl": "https://huggingface.co/Qwen/QwQ-32B",
         "websiteUrl": "https://qwenlm.github.io/blog/qwq-32b/",
@@ -262,8 +269,7 @@ envVars:
         ],
         "endpoints": [
           {
-            "type": "openai",
-            "baseURL": "https://atv7xs1nxxtx2wl0.us-east-1.aws.endpoints.huggingface.cloud/v1"
+            "type": "inference-client",
           }
         ]
       },
@@ -271,6 +277,8 @@ envVars:
         "name": "google/gemma-3-27b-it",
         "logoUrl": "https://huggingface.co/datasets/huggingchat/models-logo/resolve/main/google-logo.png",
         "multimodal": true,
+        "tools": true,
+        "systemRoleSupported" : false,
         "description": "Google's latest open model with great multilingual performance, supports image inputs natively.",
         "websiteUrl": "https://blog.google/technology/developers/gemma-3/",
         "promptExamples": [
@@ -289,15 +297,15 @@ envVars:
         ],
         "endpoints": [
           {
-            "type": "openai",
+            "type": "inference-client",
             "baseURL": "https://wp0d3hn6s3k8jk22.us-east-1.aws.endpoints.huggingface.cloud/v1",
             "multimodal": {
               "image": {
                 "maxSizeInMB": 10,
                 "maxWidth": 560,
                 "maxHeight": 560,
-                "supportedMimeTypes": ["image/jpeg"],
-                "preferredMimeType": "image/jpeg"
+                "supportedMimeTypes": ["image/jpeg", "image/png", "image/webp"],
+                "preferredMimeType": "image/webp"
               }
             }
           }
@@ -305,6 +313,7 @@ envVars:
       },
       {
         "name": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
+        "tools": true,
         "displayName": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
         "description": "A small model with good capabilities in language understanding and commonsense reasoning.",
         "logoUrl": "https://huggingface.co/datasets/huggingchat/models-logo/resolve/main/mistral-logo.png",
@@ -328,8 +337,8 @@ envVars:
 
         "endpoints": [
           {
-            "type": "openai",
-            "baseURL": "https://hkjfqcryevvq9cie.us-east-1.aws.endpoints.huggingface.cloud/v1"
+            "type": "inference-client",
+              "baseURL": "https://hkjfqcryevvq9cie.us-east-1.aws.endpoints.huggingface.cloud/v1"
           }
         ]
       },
@@ -356,8 +365,7 @@ envVars:
         ],
         "endpoints": [
           {
-            "type": "openai",
-            "baseURL": "https://lf91qeosuambouj4.us-east-1.aws.endpoints.huggingface.cloud/v1",
+            "type": "inference-client",
             "multimodal": {
               "image": {
                 "maxSizeInMB": 10,
@@ -372,6 +380,8 @@ envVars:
       },
       {
         "name": "microsoft/Phi-4",
+        "tools": true,
+        "systemRoleSupported": false,
         "description": "One of the best small models, super fast for simple tasks.",
         "logoUrl": "https://huggingface.co/datasets/huggingchat/models-logo/resolve/main/microsoft-logo.png",
         "modelUrl": "https://huggingface.co/microsoft/Phi-4",
@@ -399,7 +409,7 @@ envVars:
         ],
         "endpoints": [
           {
-            "type": "openai",
+            "type": "inference-client",
             "baseURL": "https://up5ijetg6a2e9zlb.us-east-1.aws.endpoints.huggingface.cloud/v1"
           }
         ]
@@ -407,6 +417,7 @@ envVars:
       {
         "name": "NousResearch/Hermes-3-Llama-3.1-8B",
         "description": "Nous Research's latest Hermes 3 release in 8B size. Follows instruction closely.",
+        "tools": true,
         "logoUrl": "https://huggingface.co/datasets/huggingchat/models-logo/resolve/main/nous-logo.png",
         "websiteUrl": "https://nousresearch.com/",
         "modelUrl": "https://huggingface.co/NousResearch/Hermes-3-Llama-3.1-8B",
@@ -429,23 +440,8 @@ envVars:
           "temperature": 0.6,
           "truncate": 14336,
           "max_new_tokens": 1536
-        }
-      },
-      {
-        "name": "internal/task",
-        "tokenizer" : "NousResearch/Hermes-3-Llama-3.1-8B",
-        "unlisted": true,
-        "tools" : true,
-        "endpoints": [
-          {
-            "type": "openai",
-            "baseURL": "https://api-inference.endpoints.huggingface.tech/models/NousResearch/Hermes-3-Llama-3.1-8B/v1"
-          }
-        ],
-        "parameters": {
-          "temperature": 0.1,
-          "max_new_tokens": 256
         },
+        "endpoints": [{"type" : "inference-client"}]
       }
     ]
 
@@ -539,7 +535,18 @@ envVars:
   PUBLIC_APP_DISCLAIMER: 1
   PUBLIC_PLAUSIBLE_SCRIPT_URL: "/js/script.js"
   REQUIRE_FEATURED_ASSISTANTS: "true"
-  TASK_MODEL: "internal/task"
+  TASK_MODEL: >
+        {
+        "name": "NousResearch/Hermes-3-Llama-3.1-8B",
+        "unlisted": true,
+        "endpoints": [{"type" : "inference-client"}],
+        "parameters": {
+          "temperature": 0.1,
+          "max_new_tokens": 256
+        }
+        }
+      
+
   TEXT_EMBEDDING_MODELS: >
     [{
       "name": "bge-base-en-v1-5-sxa",

package-lock.json

@@ -12,7 +12,9 @@
 				"@cliqz/adblocker-playwright": "^1.34.0",
 				"@gradio/client": "^1.8.0",
 				"@huggingface/hub": "^0.5.1",
-				"@huggingface/inference": "^2.8.1",
+				"@huggingface/inference": "^3.12.1",
+				"@huggingface/mcp-client": "^0.1.1",
+				"@huggingface/tasks": "^0.19.1",
 				"@huggingface/transformers": "^3.1.1",
 				"@iconify-json/bi": "^1.1.21",
 				"@playwright/browser-chromium": "^1.52.0",
@@ -86,6 +88,7 @@
 				"fs-extra": "^11.3.0",
 				"isomorphic-dompurify": "^2.13.0",
 				"js-yaml": "^4.1.0",
+				"jsonrepair": "^3.12.0",
 				"minimist": "^1.2.8",
 				"mongodb-memory-server": "^10.1.2",
 				"node-llama-cpp": "^3.6.0",
@@ -2032,17 +2035,27 @@
 			}
 		},
 		"node_modules/@huggingface/inference": {
-			"version": "2.8.1",
-			"resolved": "https://registry.npmjs.org/@huggingface/inference/-/inference-2.8.1.tgz",
-			"integrity": "sha512-EfsNtY9OR6JCNaUa5bZu2mrs48iqeTz0Gutwf+fU0Kypx33xFQB4DKMhp8u4Ee6qVbLbNWvTHuWwlppLQl4p4Q==",
+			"version": "3.12.1",
+			"resolved": "https://registry.npmjs.org/@huggingface/inference/-/inference-3.12.1.tgz",
+			"integrity": "sha512-gRnRHZkVAb7mzmPnLcFrttA17yZdUiz+xLqk19FBwCsCjBBlcDCXVIrZ3RW1DdZq297PB6m6t995600sDq3T6g==",
 			"license": "MIT",
 			"dependencies": {
-				"@huggingface/tasks": "^0.12.9"
+				"@huggingface/jinja": "^0.4.1",
+				"@huggingface/tasks": "^0.19.2"
 			},
 			"engines": {
 				"node": ">=18"
 			}
 		},
+		"node_modules/@huggingface/inference/node_modules/@huggingface/jinja": {
+			"version": "0.4.1",
+			"resolved": "https://registry.npmjs.org/@huggingface/jinja/-/jinja-0.4.1.tgz",
+			"integrity": "sha512-3WXbMFaPkk03LRCM0z0sylmn8ddDm4ubjU7X+Hg4M2GOuMklwoGAFXp9V2keq7vltoB/c7McE5aHUVVddAewsw==",
+			"license": "MIT",
+			"engines": {
+				"node": ">=18"
+			}
+		},
 		"node_modules/@huggingface/jinja": {
 			"version": "0.3.3",
 			"resolved": "https://registry.npmjs.org/@huggingface/jinja/-/jinja-0.3.3.tgz",
@@ -2052,10 +2065,33 @@
 				"node": ">=18"
 			}
 		},
+		"node_modules/@huggingface/mcp-client": {
+			"version": "0.1.1",
+			"resolved": "https://registry.npmjs.org/@huggingface/mcp-client/-/mcp-client-0.1.1.tgz",
+			"integrity": "sha512-jr8m0msG8XxqxJ6ibGZNcR8sHBeU3J0YCVbzjqL2UVe0YHShowMBgTdGzbuf2AVcjXbcU5bzJR+T0m7q3YooCw==",
+			"license": "MIT",
+			"dependencies": {
+				"@huggingface/inference": "^3.9.2",
+				"@huggingface/tasks": "^0.18.11",
+				"@modelcontextprotocol/sdk": "^1.9.0"
+			},
+			"bin": {
+				"tiny-agent": "dist/cli.js"
+			},
+			"engines": {
+				"node": ">=18"
+			}
+		},
+		"node_modules/@huggingface/mcp-client/node_modules/@huggingface/tasks": {
+			"version": "0.18.12",
+			"resolved": "https://registry.npmjs.org/@huggingface/tasks/-/tasks-0.18.12.tgz",
+			"integrity": "sha512-hUEYhXJzJCUhLSBNglW+Sji5fuzMnpBimp3kbdcGor+MaN9xb+qrcLnMC7X8aOcZaZdCErwYYIqz2oSP2EM4Bg==",
+			"license": "MIT"
+		},
 		"node_modules/@huggingface/tasks": {
-			"version": "0.12.30",
-			"resolved": "https://registry.npmjs.org/@huggingface/tasks/-/tasks-0.12.30.tgz",
-			"integrity": "sha512-A1ITdxbEzx9L8wKR8pF7swyrTLxWNDFIGDLUWInxvks2ruQ8PLRBZe8r0EcjC3CDdtlj9jV1V4cgV35K/iy3GQ==",
+			"version": "0.19.4",
+			"resolved": "https://registry.npmjs.org/@huggingface/tasks/-/tasks-0.19.4.tgz",
+			"integrity": "sha512-biTcoKBAtX0QaW5Rmx9O1YqU5YYpehiZDcSoAVx0SbwoMkXxFs23O59TcdPbFE4a3j5jK1eH0kltATxeGLp6dQ==",
 			"license": "MIT"
 		},
 		"node_modules/@huggingface/transformers": {
@@ -2784,6 +2820,340 @@
 			"dev": true,
 			"license": "MIT"
 		},
+		"node_modules/@modelcontextprotocol/sdk": {
+			"version": "1.10.2",
+			"resolved": "https://registry.npmjs.org/@modelcontextprotocol/sdk/-/sdk-1.10.2.tgz",
+			"integrity": "sha512-rb6AMp2DR4SN+kc6L1ta2NCpApyA9WYNx3CrTSZvGxq9wH71bRur+zRqPfg0vQ9mjywR7qZdX2RGHOPq3ss+tA==",
+			"license": "MIT",
+			"dependencies": {
+				"content-type": "^1.0.5",
+				"cors": "^2.8.5",
+				"cross-spawn": "^7.0.3",
+				"eventsource": "^3.0.2",
+				"express": "^5.0.1",
+				"express-rate-limit": "^7.5.0",
+				"pkce-challenge": "^5.0.0",
+				"raw-body": "^3.0.0",
+				"zod": "^3.23.8",
+				"zod-to-json-schema": "^3.24.1"
+			},
+			"engines": {
+				"node": ">=18"
+			}
+		},
+		"node_modules/@modelcontextprotocol/sdk/node_modules/accepts": {
+			"version": "2.0.0",
+			"resolved": "https://registry.npmjs.org/accepts/-/accepts-2.0.0.tgz",
+			"integrity": "sha512-5cvg6CtKwfgdmVqY1WIiXKc3Q1bkRqGLi+2W/6ao+6Y7gu/RCwRuAhGEzh5B4KlszSuTLgZYuqFqo5bImjNKng==",
+			"license": "MIT",
+			"dependencies": {
+				"mime-types": "^3.0.0",
+				"negotiator": "^1.0.0"
+			},
+			"engines": {
+				"node": ">= 0.6"
+			}
+		},
+		"node_modules/@modelcontextprotocol/sdk/node_modules/body-parser": {
+			"version": "2.2.0",
+			"resolved": "https://registry.npmjs.org/body-parser/-/body-parser-2.2.0.tgz",
+			"integrity": "sha512-02qvAaxv8tp7fBa/mw1ga98OGm+eCbqzJOKoRt70sLmfEEi+jyBYVTDGfCL/k06/4EMk/z01gCe7HoCH/f2LTg==",
+			"license": "MIT",
+			"dependencies": {
+				"bytes": "^3.1.2",
+				"content-type": "^1.0.5",
+				"debug": "^4.4.0",
+				"http-errors": "^2.0.0",
+				"iconv-lite": "^0.6.3",
+				"on-finished": "^2.4.1",
+				"qs": "^6.14.0",
+				"raw-body": "^3.0.0",
+				"type-is": "^2.0.0"
+			},
+			"engines": {
+				"node": ">=18"
+			}
+		},
+		"node_modules/@modelcontextprotocol/sdk/node_modules/content-disposition": {
+			"version": "1.0.0",
+			"resolved": "https://registry.npmjs.org/content-disposition/-/content-disposition-1.0.0.tgz",
+			"integrity": "sha512-Au9nRL8VNUut/XSzbQA38+M78dzP4D+eqg3gfJHMIHHYa3bg067xj1KxMUWj+VULbiZMowKngFFbKczUrNJ1mg==",
+			"license": "MIT",
+			"dependencies": {
+				"safe-buffer": "5.2.1"
+			},
+			"engines": {
+				"node": ">= 0.6"
+			}
+		},
+		"node_modules/@modelcontextprotocol/sdk/node_modules/cookie": {
+			"version": "0.7.2",
+			"resolved": "https://registry.npmjs.org/cookie/-/cookie-0.7.2.tgz",
+			"integrity": "sha512-yki5XnKuf750l50uGTllt6kKILY4nQ1eNIQatoXEByZ5dWgnKqbnqmTrBE5B4N7lrMJKQ2ytWMiTO2o0v6Ew/w==",
+			"license": "MIT",
+			"engines": {
+				"node": ">= 0.6"
+			}
+		},
+		"node_modules/@modelcontextprotocol/sdk/node_modules/cookie-signature": {
+			"version": "1.2.2",
+			"resolved": "https://registry.npmjs.org/cookie-signature/-/cookie-signature-1.2.2.tgz",
+			"integrity": "sha512-D76uU73ulSXrD1UXF4KE2TMxVVwhsnCgfAyTg9k8P6KGZjlXKrOLe4dJQKI3Bxi5wjesZoFXJWElNWBjPZMbhg==",
+			"license": "MIT",
+			"engines": {
+				"node": ">=6.6.0"
+			}
+		},
+		"node_modules/@modelcontextprotocol/sdk/node_modules/debug": {
+			"version": "4.4.0",
+			"resolved": "https://registry.npmjs.org/debug/-/debug-4.4.0.tgz",
+			"integrity": "sha512-6WTZ/IxCY/T6BALoZHaE4ctp9xm+Z5kY/pzYaCHRFeyVhojxlrm+46y68HA6hr0TcwEssoxNiDEUJQjfPZ/RYA==",
+			"license": "MIT",
+			"dependencies": {
+				"ms": "^2.1.3"
+			},
+			"engines": {
+				"node": ">=6.0"
+			},
+			"peerDependenciesMeta": {
+				"supports-color": {
+					"optional": true
+				}
+			}
+		},
+		"node_modules/@modelcontextprotocol/sdk/node_modules/encodeurl": {
+			"version": "2.0.0",
+			"resolved": "https://registry.npmjs.org/encodeurl/-/encodeurl-2.0.0.tgz",
+			"integrity": "sha512-Q0n9HRi4m6JuGIV1eFlmvJB7ZEVxu93IrMyiMsGC0lrMJMWzRgx6WGquyfQgZVb31vhGgXnfmPNNXmxnOkRBrg==",
+			"license": "MIT",
+			"engines": {
+				"node": ">= 0.8"
+			}
+		},
+		"node_modules/@modelcontextprotocol/sdk/node_modules/eventsource": {
+			"version": "3.0.6",
+			"resolved": "https://registry.npmjs.org/eventsource/-/eventsource-3.0.6.tgz",
+			"integrity": "sha512-l19WpE2m9hSuyP06+FbuUUf1G+R0SFLrtQfbRb9PRr+oimOfxQhgGCbVaXg5IvZyyTThJsxh6L/srkMiCeBPDA==",
+			"license": "MIT",
+			"dependencies": {
+				"eventsource-parser": "^3.0.1"
+			},
+			"engines": {
+				"node": ">=18.0.0"
+			}
+		},
+		"node_modules/@modelcontextprotocol/sdk/node_modules/express": {
+			"version": "5.1.0",
+			"resolved": "https://registry.npmjs.org/express/-/express-5.1.0.tgz",
+			"integrity": "sha512-DT9ck5YIRU+8GYzzU5kT3eHGA5iL+1Zd0EutOmTE9Dtk+Tvuzd23VBU+ec7HPNSTxXYO55gPV/hq4pSBJDjFpA==",
+			"license": "MIT",
+			"dependencies": {
+				"accepts": "^2.0.0",
+				"body-parser": "^2.2.0",
+				"content-disposition": "^1.0.0",
+				"content-type": "^1.0.5",
+				"cookie": "^0.7.1",
+				"cookie-signature": "^1.2.1",
+				"debug": "^4.4.0",
+				"encodeurl": "^2.0.0",
+				"escape-html": "^1.0.3",
+				"etag": "^1.8.1",
+				"finalhandler": "^2.1.0",
+				"fresh": "^2.0.0",
+				"http-errors": "^2.0.0",
+				"merge-descriptors": "^2.0.0",
+				"mime-types": "^3.0.0",
+				"on-finished": "^2.4.1",
+				"once": "^1.4.0",
+				"parseurl": "^1.3.3",
+				"proxy-addr": "^2.0.7",
+				"qs": "^6.14.0",
+				"range-parser": "^1.2.1",
+				"router": "^2.2.0",
+				"send": "^1.1.0",
+				"serve-static": "^2.2.0",
+				"statuses": "^2.0.1",
+				"type-is": "^2.0.1",
+				"vary": "^1.1.2"
+			},
+			"engines": {
+				"node": ">= 18"
+			},
+			"funding": {
+				"type": "opencollective",
+				"url": "https://opencollective.com/express"
+			}
+		},
+		"node_modules/@modelcontextprotocol/sdk/node_modules/finalhandler": {
+			"version": "2.1.0",
+			"resolved": "https://registry.npmjs.org/finalhandler/-/finalhandler-2.1.0.tgz",
+			"integrity": "sha512-/t88Ty3d5JWQbWYgaOGCCYfXRwV1+be02WqYYlL6h0lEiUAMPM8o8qKGO01YIkOHzka2up08wvgYD0mDiI+q3Q==",
+			"license": "MIT",
+			"dependencies": {
+				"debug": "^4.4.0",
+				"encodeurl": "^2.0.0",
+				"escape-html": "^1.0.3",
+				"on-finished": "^2.4.1",
+				"parseurl": "^1.3.3",
+				"statuses": "^2.0.1"
+			},
+			"engines": {
+				"node": ">= 0.8"
+			}
+		},
+		"node_modules/@modelcontextprotocol/sdk/node_modules/fresh": {
+			"version": "2.0.0",
+			"resolved": "https://registry.npmjs.org/fresh/-/fresh-2.0.0.tgz",
+			"integrity": "sha512-Rx/WycZ60HOaqLKAi6cHRKKI7zxWbJ31MhntmtwMoaTeF7XFH9hhBp8vITaMidfljRQ6eYWCKkaTK+ykVJHP2A==",
+			"license": "MIT",
+			"engines": {
+				"node": ">= 0.8"
+			}
+		},
+		"node_modules/@modelcontextprotocol/sdk/node_modules/iconv-lite": {
+			"version": "0.6.3",
+			"resolved": "https://registry.npmjs.org/iconv-lite/-/iconv-lite-0.6.3.tgz",
+			"integrity": "sha512-4fCk79wshMdzMp2rH06qWrJE4iolqLhCUH+OiuIgU++RB0+94NlDL81atO7GX55uUKueo0txHNtvEyI6D7WdMw==",
+			"license": "MIT",
+			"dependencies": {
+				"safer-buffer": ">= 2.1.2 < 3.0.0"
+			},
+			"engines": {
+				"node": ">=0.10.0"
+			}
+		},
+		"node_modules/@modelcontextprotocol/sdk/node_modules/media-typer": {
+			"version": "1.1.0",
+			"resolved": "https://registry.npmjs.org/media-typer/-/media-typer-1.1.0.tgz",
+			"integrity": "sha512-aisnrDP4GNe06UcKFnV5bfMNPBUw4jsLGaWwWfnH3v02GnBuXX2MCVn5RbrWo0j3pczUilYblq7fQ7Nw2t5XKw==",
+			"license": "MIT",
+			"engines": {
+				"node": ">= 0.8"
+			}
+		},
+		"node_modules/@modelcontextprotocol/sdk/node_modules/merge-descriptors": {
+			"version": "2.0.0",
+			"resolved": "https://registry.npmjs.org/merge-descriptors/-/merge-descriptors-2.0.0.tgz",
+			"integrity": "sha512-Snk314V5ayFLhp3fkUREub6WtjBfPdCPY1Ln8/8munuLuiYhsABgBVWsozAG+MWMbVEvcdcpbi9R7ww22l9Q3g==",
+			"license": "MIT",
+			"engines": {
+				"node": ">=18"
+			},
+			"funding": {
+				"url": "https://github.com/sponsors/sindresorhus"
+			}
+		},
+		"node_modules/@modelcontextprotocol/sdk/node_modules/mime-db": {
+			"version": "1.54.0",
+			"resolved": "https://registry.npmjs.org/mime-db/-/mime-db-1.54.0.tgz",
+			"integrity": "sha512-aU5EJuIN2WDemCcAp2vFBfp/m4EAhWJnUNSSw0ixs7/kXbd6Pg64EmwJkNdFhB8aWt1sH2CTXrLxo/iAGV3oPQ==",
+			"license": "MIT",
+			"engines": {
+				"node": ">= 0.6"
+			}
+		},
+		"node_modules/@modelcontextprotocol/sdk/node_modules/mime-types": {
+			"version": "3.0.1",
+			"resolved": "https://registry.npmjs.org/mime-types/-/mime-types-3.0.1.tgz",
+			"integrity": "sha512-xRc4oEhT6eaBpU1XF7AjpOFD+xQmXNB5OVKwp4tqCuBpHLS/ZbBDrc07mYTDqVMg6PfxUjjNp85O6Cd2Z/5HWA==",
+			"license": "MIT",
+			"dependencies": {
+				"mime-db": "^1.54.0"
+			},
+			"engines": {
+				"node": ">= 0.6"
+			}
+		},
+		"node_modules/@modelcontextprotocol/sdk/node_modules/negotiator": {
+			"version": "1.0.0",
+			"resolved": "https://registry.npmjs.org/negotiator/-/negotiator-1.0.0.tgz",
+			"integrity": "sha512-8Ofs/AUQh8MaEcrlq5xOX0CQ9ypTF5dl78mjlMNfOK08fzpgTHQRQPBxcPlEtIw0yRpws+Zo/3r+5WRby7u3Gg==",
+			"license": "MIT",
+			"engines": {
+				"node": ">= 0.6"
+			}
+		},
+		"node_modules/@modelcontextprotocol/sdk/node_modules/qs": {
+			"version": "6.14.0",
+			"resolved": "https://registry.npmjs.org/qs/-/qs-6.14.0.tgz",
+			"integrity": "sha512-YWWTjgABSKcvs/nWBi9PycY/JiPJqOD4JA6o9Sej2AtvSGarXxKC3OQSk4pAarbdQlKAh5D4FCQkJNkW+GAn3w==",
+			"license": "BSD-3-Clause",
+			"dependencies": {
+				"side-channel": "^1.1.0"
+			},
+			"engines": {
+				"node": ">=0.6"
+			},
+			"funding": {
+				"url": "https://github.com/sponsors/ljharb"
+			}
+		},
+		"node_modules/@modelcontextprotocol/sdk/node_modules/raw-body": {
+			"version": "3.0.0",
+			"resolved": "https://registry.npmjs.org/raw-body/-/raw-body-3.0.0.tgz",
+			"integrity": "sha512-RmkhL8CAyCRPXCE28MMH0z2PNWQBNk2Q09ZdxM9IOOXwxwZbN+qbWaatPkdkWIKL2ZVDImrN/pK5HTRz2PcS4g==",
+			"license": "MIT",
+			"dependencies": {
+				"bytes": "3.1.2",
+				"http-errors": "2.0.0",
+				"iconv-lite": "0.6.3",
+				"unpipe": "1.0.0"
+			},
+			"engines": {
+				"node": ">= 0.8"
+			}
+		},
+		"node_modules/@modelcontextprotocol/sdk/node_modules/send": {
+			"version": "1.2.0",
+			"resolved": "https://registry.npmjs.org/send/-/send-1.2.0.tgz",
+			"integrity": "sha512-uaW0WwXKpL9blXE2o0bRhoL2EGXIrZxQ2ZQ4mgcfoBxdFmQold+qWsD2jLrfZ0trjKL6vOw0j//eAwcALFjKSw==",
+			"license": "MIT",
+			"dependencies": {
+				"debug": "^4.3.5",
+				"encodeurl": "^2.0.0",
+				"escape-html": "^1.0.3",
+				"etag": "^1.8.1",
+				"fresh": "^2.0.0",
+				"http-errors": "^2.0.0",
+				"mime-types": "^3.0.1",
+				"ms": "^2.1.3",
+				"on-finished": "^2.4.1",
+				"range-parser": "^1.2.1",
+				"statuses": "^2.0.1"
+			},
+			"engines": {
+				"node": ">= 18"
+			}
+		},
+		"node_modules/@modelcontextprotocol/sdk/node_modules/serve-static": {
+			"version": "2.2.0",
+			"resolved": "https://registry.npmjs.org/serve-static/-/serve-static-2.2.0.tgz",
+			"integrity": "sha512-61g9pCh0Vnh7IutZjtLGGpTA355+OPn2TyDv/6ivP2h/AdAVX9azsoxmg2/M6nZeQZNYBEwIcsne1mJd9oQItQ==",
+			"license": "MIT",
+			"dependencies": {
+				"encodeurl": "^2.0.0",
+				"escape-html": "^1.0.3",
+				"parseurl": "^1.3.3",
+				"send": "^1.2.0"
+			},
+			"engines": {
+				"node": ">= 18"
+			}
+		},
+		"node_modules/@modelcontextprotocol/sdk/node_modules/type-is": {
+			"version": "2.0.1",
+			"resolved": "https://registry.npmjs.org/type-is/-/type-is-2.0.1.tgz",
+			"integrity": "sha512-OZs6gsjF4vMp32qrCbiVSkrFmXtG/AZhY3t0iAMrMBiAZyV9oALtXO8hsrHbMXF9x6L3grlFuwW2oAz7cav+Gw==",
+			"license": "MIT",
+			"dependencies": {
+				"content-type": "^1.0.5",
+				"media-typer": "^1.1.0",
+				"mime-types": "^3.0.0"
+			},
+			"engines": {
+				"node": ">= 0.6"
+			}
+		},
 		"node_modules/@mongodb-js/saslprep": {
 			"version": "1.1.9",
 			"resolved": "https://registry.npmjs.org/@mongodb-js/saslprep/-/saslprep-1.1.9.tgz",
@@ -6264,16 +6634,27 @@
 				"node": ">=8"
 			}
 		},
-		"node_modules/call-bind": {
-			"version": "1.0.7",
-			"resolved": "https://registry.npmjs.org/call-bind/-/call-bind-1.0.7.tgz",
-			"integrity": "sha512-GHTSNSYICQ7scH7sZ+M2rFopRoLh8t2bLSW6BbgrtLsahOIB5iyAVJf9GjWK3cYTDaMj4XdBpM1cA6pIS0Kv2w==",
+		"node_modules/call-bind-apply-helpers": {
+			"version": "1.0.2",
+			"resolved": "https://registry.npmjs.org/call-bind-apply-helpers/-/call-bind-apply-helpers-1.0.2.tgz",
+			"integrity": "sha512-Sp1ablJ0ivDkSzjcaJdxEunN5/XvksFJ2sMBFfq6x0ryhQV/2b/KwFe21cMpmHtPOSij8K99/wSfoEuTObmuMQ==",
+			"license": "MIT",
 			"dependencies": {
-				"es-define-property": "^1.0.0",
 				"es-errors": "^1.3.0",
-				"function-bind": "^1.1.2",
-				"get-intrinsic": "^1.2.4",
-				"set-function-length": "^1.2.1"
+				"function-bind": "^1.1.2"
+			},
+			"engines": {
+				"node": ">= 0.4"
+			}
+		},
+		"node_modules/call-bound": {
+			"version": "1.0.4",
+			"resolved": "https://registry.npmjs.org/call-bound/-/call-bound-1.0.4.tgz",
+			"integrity": "sha512-+ys997U96po4Kx/ABpBCqhA9EuxJaQWDQg7295H4hBphv3IZg0boBKuwYpt4YXp6MZ5AmZQnU/tyMTlRpaSejg==",
+			"license": "MIT",
+			"dependencies": {
+				"call-bind-apply-helpers": "^1.0.2",
+				"get-intrinsic": "^1.3.0"
 			},
 			"engines": {
 				"node": ">= 0.4"
@@ -6866,6 +7247,19 @@
 			"resolved": "https://registry.npmjs.org/cookie-signature/-/cookie-signature-1.0.6.tgz",
 			"integrity": "sha512-QADzlaHc8icV8I7vbaJXJwod9HWYp8uCqf1xa4OfNu1T7JVxQIrUgOWtHdNDtPiywmFbiS12VjotIXLrKM3orQ=="
 		},
+		"node_modules/cors": {
+			"version": "2.8.5",
+			"resolved": "https://registry.npmjs.org/cors/-/cors-2.8.5.tgz",
+			"integrity": "sha512-KIHbLJqu73RGr/hnbrO9uBeixNGuvSQjul/jdFvS/KFSIH1hWVd1ng7zOHx+YrEfInLG7q4n6GHQ9cDtxv/P6g==",
+			"license": "MIT",
+			"dependencies": {
+				"object-assign": "^4",
+				"vary": "^1"
+			},
+			"engines": {
+				"node": ">= 0.10"
+			}
+		},
 		"node_modules/create-require": {
 			"version": "1.1.1",
 			"resolved": "https://registry.npmjs.org/create-require/-/create-require-1.1.1.tgz",
@@ -7045,22 +7439,6 @@
 				"node": ">=0.10.0"
 			}
 		},
-		"node_modules/define-data-property": {
-			"version": "1.1.4",
-			"resolved": "https://registry.npmjs.org/define-data-property/-/define-data-property-1.1.4.tgz",
-			"integrity": "sha512-rBMvIzlpA8v6E+SJZoo++HAYqsLrkg7MSfIinMPFhmkorw7X+dOXVJQs+QT69zGkzMyfDnIMN2Wid1+NbL3T+A==",
-			"dependencies": {
-				"es-define-property": "^1.0.0",
-				"es-errors": "^1.3.0",
-				"gopd": "^1.0.1"
-			},
-			"engines": {
-				"node": ">= 0.4"
-			},
-			"funding": {
-				"url": "https://github.com/sponsors/ljharb"
-			}
-		},
 		"node_modules/delayed-stream": {
 			"version": "1.0.0",
 			"resolved": "https://registry.npmjs.org/delayed-stream/-/delayed-stream-1.0.0.tgz",
@@ -7235,6 +7613,20 @@
 				"url": "https://dotenvx.com"
 			}
 		},
+		"node_modules/dunder-proto": {
+			"version": "1.0.1",
+			"resolved": "https://registry.npmjs.org/dunder-proto/-/dunder-proto-1.0.1.tgz",
+			"integrity": "sha512-KIN/nDJBQRcXw0MLVhZE9iQHmG68qAVIBg9CqmUYjmQIhgij9U5MFvrqkUL5FbtyyzZuOeOt0zdeRe4UY7ct+A==",
+			"license": "MIT",
+			"dependencies": {
+				"call-bind-apply-helpers": "^1.0.1",
+				"es-errors": "^1.3.0",
+				"gopd": "^1.2.0"
+			},
+			"engines": {
+				"node": ">= 0.4"
+			}
+		},
 		"node_modules/eastasianwidth": {
 			"version": "0.2.0",
 			"resolved": "https://registry.npmjs.org/eastasianwidth/-/eastasianwidth-0.2.0.tgz",
@@ -7312,12 +7704,10 @@
 			}
 		},
 		"node_modules/es-define-property": {
-			"version": "1.0.0",
-			"resolved": "https://registry.npmjs.org/es-define-property/-/es-define-property-1.0.0.tgz",
-			"integrity": "sha512-jxayLKShrEqqzJ0eumQbVhTYQM27CfT1T35+gCgDFoL82JLsXqTJ76zv6A0YLOgEnLUMvLzsDsGIrl8NFpT2gQ==",
-			"dependencies": {
-				"get-intrinsic": "^1.2.4"
-			},
+			"version": "1.0.1",
+			"resolved": "https://registry.npmjs.org/es-define-property/-/es-define-property-1.0.1.tgz",
+			"integrity": "sha512-e3nRfgfUZ4rNGL232gUgX06QNyyez04KdjFrF+LTRoOXmrOgFKDg4BCdsjW8EnT69eqdYGmRpJwiPVYNrCaW3g==",
+			"license": "MIT",
 			"engines": {
 				"node": ">= 0.4"
 			}
@@ -7326,6 +7716,7 @@
 			"version": "1.3.0",
 			"resolved": "https://registry.npmjs.org/es-errors/-/es-errors-1.3.0.tgz",
 			"integrity": "sha512-Zf5H2Kxt2xjTvbJvP2ZWLEICxA6j+hAmMzIlypy4xcBg1vKVnx89Wy0GbS+kf5cwCVFFzdCFh2XSCFNULS6csw==",
+			"license": "MIT",
 			"engines": {
 				"node": ">= 0.4"
 			}
@@ -7337,6 +7728,18 @@
 			"dev": true,
 			"license": "MIT"
 		},
+		"node_modules/es-object-atoms": {
+			"version": "1.1.1",
+			"resolved": "https://registry.npmjs.org/es-object-atoms/-/es-object-atoms-1.1.1.tgz",
+			"integrity": "sha512-FGgH2h8zKNim9ljj7dankFPcICIK9Cp5bm+c2gQSYePhpaG5+esrLODihIorn+Pe6FGJzWhXQotPv73jTaldXA==",
+			"license": "MIT",
+			"dependencies": {
+				"es-errors": "^1.3.0"
+			},
+			"engines": {
+				"node": ">= 0.4"
+			}
+		},
 		"node_modules/esbuild": {
 			"version": "0.25.1",
 			"resolved": "https://registry.npmjs.org/esbuild/-/esbuild-0.25.1.tgz",
@@ -7800,6 +8203,15 @@
 				"node": ">=12.0.0"
 			}
 		},
+		"node_modules/eventsource-parser": {
+			"version": "3.0.1",
+			"resolved": "https://registry.npmjs.org/eventsource-parser/-/eventsource-parser-3.0.1.tgz",
+			"integrity": "sha512-VARTJ9CYeuQYb0pZEPbzi740OWFgpHe7AYJ2WFZVnUDUQp5Dk2yJUgF36YsZ81cOyxT0QxmXD2EQpapAouzWVA==",
+			"license": "MIT",
+			"engines": {
+				"node": ">=18.0.0"
+			}
+		},
 		"node_modules/execa": {
 			"version": "8.0.1",
 			"resolved": "https://registry.npmjs.org/execa/-/execa-8.0.1.tgz",
@@ -7899,6 +8311,21 @@
 				"url": "https://opencollective.com/express"
 			}
 		},
+		"node_modules/express-rate-limit": {
+			"version": "7.5.0",
+			"resolved": "https://registry.npmjs.org/express-rate-limit/-/express-rate-limit-7.5.0.tgz",
+			"integrity": "sha512-eB5zbQh5h+VenMPM3fh+nw1YExi5nMr6HUCR62ELSP11huvxm/Uir1H1QEyTkk5QX6A58pX6NmaTMceKZ0Eodg==",
+			"license": "MIT",
+			"engines": {
+				"node": ">= 16"
+			},
+			"funding": {
+				"url": "https://github.com/sponsors/express-rate-limit"
+			},
+			"peerDependencies": {
+				"express": "^4.11 || 5 || ^5.0.0-beta.1"
+			}
+		},
 		"node_modules/express/node_modules/cookie": {
 			"version": "0.7.1",
 			"resolved": "https://registry.npmjs.org/cookie/-/cookie-0.7.1.tgz",
@@ -8542,15 +8969,21 @@
 			}
 		},
 		"node_modules/get-intrinsic": {
-			"version": "1.2.4",
-			"resolved": "https://registry.npmjs.org/get-intrinsic/-/get-intrinsic-1.2.4.tgz",
-			"integrity": "sha512-5uYhsJH8VJBTv7oslg4BznJYhDoRI6waYCxMmCdnTrcCrHA/fCFKoTFz2JKKE0HdDFUF7/oQuhzumXJK7paBRQ==",
+			"version": "1.3.0",
+			"resolved": "https://registry.npmjs.org/get-intrinsic/-/get-intrinsic-1.3.0.tgz",
+			"integrity": "sha512-9fSjSaos/fRIVIp+xSJlE6lfwhES7LNtKaCBIamHsjr2na1BiABJPo0mOjjz8GJDURarmCPGqaiVg5mfjb98CQ==",
+			"license": "MIT",
 			"dependencies": {
+				"call-bind-apply-helpers": "^1.0.2",
+				"es-define-property": "^1.0.1",
 				"es-errors": "^1.3.0",
+				"es-object-atoms": "^1.1.1",
 				"function-bind": "^1.1.2",
-				"has-proto": "^1.0.1",
-				"has-symbols": "^1.0.3",
-				"hasown": "^2.0.0"
+				"get-proto": "^1.0.1",
+				"gopd": "^1.2.0",
+				"has-symbols": "^1.1.0",
+				"hasown": "^2.0.2",
+				"math-intrinsics": "^1.1.0"
 			},
 			"engines": {
 				"node": ">= 0.4"
@@ -8559,6 +8992,19 @@
 				"url": "https://github.com/sponsors/ljharb"
 			}
 		},
+		"node_modules/get-proto": {
+			"version": "1.0.1",
+			"resolved": "https://registry.npmjs.org/get-proto/-/get-proto-1.0.1.tgz",
+			"integrity": "sha512-sTSfBjoXBp89JvIKIefqw7U2CCebsc74kiY6awiGogKtoSGbgjYE/G/+l9sF3MWFPNc9IcoOC4ODfKHfxFmp0g==",
+			"license": "MIT",
+			"dependencies": {
+				"dunder-proto": "^1.0.1",
+				"es-object-atoms": "^1.0.0"
+			},
+			"engines": {
+				"node": ">= 0.4"
+			}
+		},
 		"node_modules/get-stream": {
 			"version": "9.0.1",
 			"resolved": "https://registry.npmjs.org/get-stream/-/get-stream-9.0.1.tgz",
@@ -8670,11 +9116,12 @@
 			}
 		},
 		"node_modules/gopd": {
-			"version": "1.0.1",
-			"resolved": "https://registry.npmjs.org/gopd/-/gopd-1.0.1.tgz",
-			"integrity": "sha512-d65bNlIadxvpb/A2abVdlqKqV563juRnZ1Wtk6s1sIR8uNsXR70xqIzVqxVf1eTqDunwT2MkczEeaezCKTZhwA==",
-			"dependencies": {
-				"get-intrinsic": "^1.1.3"
+			"version": "1.2.0",
+			"resolved": "https://registry.npmjs.org/gopd/-/gopd-1.2.0.tgz",
+			"integrity": "sha512-ZUKRh6/kUFoAiTAtTYPZJ3hw9wNxx+BIBOijnlG9PnrJsCcSjs1wyyD6vJpaYtgnzDrKYRSqf3OO6Rfa93xsRg==",
+			"license": "MIT",
+			"engines": {
+				"node": ">= 0.4"
 			},
 			"funding": {
 				"url": "https://github.com/sponsors/ljharb"
@@ -8747,32 +9194,11 @@
 				"node": ">=8"
 			}
 		},
-		"node_modules/has-property-descriptors": {
-			"version": "1.0.2",
-			"resolved": "https://registry.npmjs.org/has-property-descriptors/-/has-property-descriptors-1.0.2.tgz",
-			"integrity": "sha512-55JNKuIW+vq4Ke1BjOTjM2YctQIvCT7GFzHwmfZPGo5wnrgkid0YQtnAleFSqumZm4az3n2BS+erby5ipJdgrg==",
-			"dependencies": {
-				"es-define-property": "^1.0.0"
-			},
-			"funding": {
-				"url": "https://github.com/sponsors/ljharb"
-			}
-		},
-		"node_modules/has-proto": {
-			"version": "1.0.3",
-			"resolved": "https://registry.npmjs.org/has-proto/-/has-proto-1.0.3.tgz",
-			"integrity": "sha512-SJ1amZAJUiZS+PhsVLf5tGydlaVB8EdFpaSO4gmiUKUOxk8qzn5AIy4ZeJUmh22znIdk/uMAUT2pl3FxzVUH+Q==",
-			"engines": {
-				"node": ">= 0.4"
-			},
-			"funding": {
-				"url": "https://github.com/sponsors/ljharb"
-			}
-		},
 		"node_modules/has-symbols": {
-			"version": "1.0.3",
-			"resolved": "https://registry.npmjs.org/has-symbols/-/has-symbols-1.0.3.tgz",
-			"integrity": "sha512-l3LCuF6MgDNwTDKkdYGEihYjt5pRPbEg46rtlmnSPlUbgmB8LOIrKJbYYFBSbnPaJexMKtiPO8hmeRjRz2Td+A==",
+			"version": "1.1.0",
+			"resolved": "https://registry.npmjs.org/has-symbols/-/has-symbols-1.1.0.tgz",
+			"integrity": "sha512-1cDNdwJ2Jaohmb3sg4OmKaMBwuC48sYni5HUw2DvsC8LjGTLK9h+eb1X6RyuOHe4hT0ULCW68iomhjUoKUqlPQ==",
+			"license": "MIT",
 			"engines": {
 				"node": ">= 0.4"
 			},
@@ -9375,6 +9801,12 @@
 			"resolved": "https://registry.npmjs.org/is-potential-custom-element-name/-/is-potential-custom-element-name-1.0.1.tgz",
 			"integrity": "sha512-bCYeRA2rVibKZd+s2625gGnGF/t7DSqDs4dP7CrLA1m7jKWz6pps0LpYLJN8Q64HtmPKJ1hrN3nzPNKFEKOUiQ=="
 		},
+		"node_modules/is-promise": {
+			"version": "4.0.0",
+			"resolved": "https://registry.npmjs.org/is-promise/-/is-promise-4.0.0.tgz",
+			"integrity": "sha512-hvpoI6korhJMnej285dSg6nu1+e6uxs7zG3BYAm5byqDsgJNWwxzM6z6iZiAgQR4TJ30JmBTOwqZUw3WlyH3AQ==",
+			"license": "MIT"
+		},
 		"node_modules/is-reference": {
 			"version": "1.2.1",
 			"resolved": "https://registry.npmjs.org/is-reference/-/is-reference-1.2.1.tgz",
@@ -9798,6 +10230,16 @@
 				"underscore": "1.12.1"
 			}
 		},
+		"node_modules/jsonrepair": {
+			"version": "3.12.0",
+			"resolved": "https://registry.npmjs.org/jsonrepair/-/jsonrepair-3.12.0.tgz",
+			"integrity": "sha512-SWfjz8SuQ0wZjwsxtSJ3Zy8vvLg6aO/kxcp9TWNPGwJKgTZVfhNEQBMk/vPOpYCDFWRxD6QWuI6IHR1t615f0w==",
+			"dev": true,
+			"license": "ISC",
+			"bin": {
+				"jsonrepair": "bin/cli.js"
+			}
+		},
 		"node_modules/jwa": {
 			"version": "2.0.0",
 			"resolved": "https://registry.npmjs.org/jwa/-/jwa-2.0.0.tgz",
@@ -10338,6 +10780,15 @@
 				"node": ">= 18"
 			}
 		},
+		"node_modules/math-intrinsics": {
+			"version": "1.1.0",
+			"resolved": "https://registry.npmjs.org/math-intrinsics/-/math-intrinsics-1.1.0.tgz",
+			"integrity": "sha512-/IXtbwEk5HTPyEwyKX6hGkYXxM9nbj64B+ilVJnC/R6B0pH5G4V3b0pVbL7DBj4tkhBAppbQUlf6F6Xl9LHu1g==",
+			"license": "MIT",
+			"engines": {
+				"node": ">= 0.4"
+			}
+		},
 		"node_modules/media-typer": {
 			"version": "0.3.0",
 			"resolved": "https://registry.npmjs.org/media-typer/-/media-typer-0.3.0.tgz",
@@ -11359,9 +11810,10 @@
 			}
 		},
 		"node_modules/object-inspect": {
-			"version": "1.13.2",
-			"resolved": "https://registry.npmjs.org/object-inspect/-/object-inspect-1.13.2.tgz",
-			"integrity": "sha512-IRZSRuzJiynemAXPYtPe5BoI/RESNYR7TYm50MC5Mqbd3Jmw5y790sErYw3V6SryFJD64b74qQQs9wn5Bg/k3g==",
+			"version": "1.13.4",
+			"resolved": "https://registry.npmjs.org/object-inspect/-/object-inspect-1.13.4.tgz",
+			"integrity": "sha512-W67iLl4J2EXEGTbfeHCffrjDfitvLANg0UlX3wFUUSTx92KXRFegMHUVgSqE+wvhAbi4WqjGg9czysTV2Epbew==",
+			"license": "MIT",
 			"engines": {
 				"node": ">= 0.4"
 			},
@@ -12031,6 +12483,15 @@
 				"node": ">= 6"
 			}
 		},
+		"node_modules/pkce-challenge": {
+			"version": "5.0.0",
+			"resolved": "https://registry.npmjs.org/pkce-challenge/-/pkce-challenge-5.0.0.tgz",
+			"integrity": "sha512-ueGLflrrnvwB3xuo/uGob5pd5FN7l0MsLf0Z87o/UQmRtwjvfylfc9MurIxRAWywCYTgrvpXBcqjV4OfCYGCIQ==",
+			"license": "MIT",
+			"engines": {
+				"node": ">=16.20.0"
+			}
+		},
 		"node_modules/pkg-dir": {
 			"version": "4.2.0",
 			"resolved": "https://registry.npmjs.org/pkg-dir/-/pkg-dir-4.2.0.tgz",
@@ -12989,6 +13450,48 @@
 				"fsevents": "~2.3.2"
 			}
 		},
+		"node_modules/router": {
+			"version": "2.2.0",
+			"resolved": "https://registry.npmjs.org/router/-/router-2.2.0.tgz",
+			"integrity": "sha512-nLTrUKm2UyiL7rlhapu/Zl45FwNgkZGaCpZbIHajDYgwlJCOzLSk+cIPAnsEqV955GjILJnKbdQC1nVPz+gAYQ==",
+			"license": "MIT",
+			"dependencies": {
+				"debug": "^4.4.0",
+				"depd": "^2.0.0",
+				"is-promise": "^4.0.0",
+				"parseurl": "^1.3.3",
+				"path-to-regexp": "^8.0.0"
+			},
+			"engines": {
+				"node": ">= 18"
+			}
+		},
+		"node_modules/router/node_modules/debug": {
+			"version": "4.4.0",
+			"resolved": "https://registry.npmjs.org/debug/-/debug-4.4.0.tgz",
+			"integrity": "sha512-6WTZ/IxCY/T6BALoZHaE4ctp9xm+Z5kY/pzYaCHRFeyVhojxlrm+46y68HA6hr0TcwEssoxNiDEUJQjfPZ/RYA==",
+			"license": "MIT",
+			"dependencies": {
+				"ms": "^2.1.3"
+			},
+			"engines": {
+				"node": ">=6.0"
+			},
+			"peerDependenciesMeta": {
+				"supports-color": {
+					"optional": true
+				}
+			}
+		},
+		"node_modules/router/node_modules/path-to-regexp": {
+			"version": "8.2.0",
+			"resolved": "https://registry.npmjs.org/path-to-regexp/-/path-to-regexp-8.2.0.tgz",
+			"integrity": "sha512-TdrF7fW9Rphjq4RjrW0Kp2AW0Ahwu9sRGTkS6bvDi0SCwZlEZYmcfDbEsTz8RVk0EHIS/Vd1bv3JhG+1xZuAyQ==",
+			"license": "MIT",
+			"engines": {
+				"node": ">=16"
+			}
+		},
 		"node_modules/rrweb-cssom": {
 			"version": "0.6.0",
 			"resolved": "https://registry.npmjs.org/rrweb-cssom/-/rrweb-cssom-0.6.0.tgz",
@@ -13236,22 +13739,6 @@
 			"integrity": "sha512-lXLOiqpkUumhRdFF3k1osNXCy9akgx/dyPZ5p8qAg9seJzXr5ZrlqZuWIMuY6ejOsVLE6flJ5/h3lsn57fQ/PQ==",
 			"dev": true
 		},
-		"node_modules/set-function-length": {
-			"version": "1.2.2",
-			"resolved": "https://registry.npmjs.org/set-function-length/-/set-function-length-1.2.2.tgz",
-			"integrity": "sha512-pgRc4hJ4/sNjWCSS9AmnS40x3bNMDTknHgL5UaMBTMyJnU90EgWh1Rz+MC9eFu4BuN/UwZjKQuY/1v3rM7HMfg==",
-			"dependencies": {
-				"define-data-property": "^1.1.4",
-				"es-errors": "^1.3.0",
-				"function-bind": "^1.1.2",
-				"get-intrinsic": "^1.2.4",
-				"gopd": "^1.0.1",
-				"has-property-descriptors": "^1.0.2"
-			},
-			"engines": {
-				"node": ">= 0.4"
-			}
-		},
 		"node_modules/setprototypeof": {
 			"version": "1.2.0",
 			"resolved": "https://registry.npmjs.org/setprototypeof/-/setprototypeof-1.2.0.tgz",
@@ -13315,14 +13802,69 @@
 			}
 		},
 		"node_modules/side-channel": {
-			"version": "1.0.6",
-			"resolved": "https://registry.npmjs.org/side-channel/-/side-channel-1.0.6.tgz",
-			"integrity": "sha512-fDW/EZ6Q9RiO8eFG8Hj+7u/oW+XrPTIChwCOM2+th2A6OblDtYYIpve9m+KvI9Z4C9qSEXlaGR6bTEYHReuglA==",
+			"version": "1.1.0",
+			"resolved": "https://registry.npmjs.org/side-channel/-/side-channel-1.1.0.tgz",
+			"integrity": "sha512-ZX99e6tRweoUXqR+VBrslhda51Nh5MTQwou5tnUDgbtyM0dBgmhEDtWGP/xbKn6hqfPRHujUNwz5fy/wbbhnpw==",
+			"license": "MIT",
+			"dependencies": {
+				"es-errors": "^1.3.0",
+				"object-inspect": "^1.13.3",
+				"side-channel-list": "^1.0.0",
+				"side-channel-map": "^1.0.1",
+				"side-channel-weakmap": "^1.0.2"
+			},
+			"engines": {
+				"node": ">= 0.4"
+			},
+			"funding": {
+				"url": "https://github.com/sponsors/ljharb"
+			}
+		},
+		"node_modules/side-channel-list": {
+			"version": "1.0.0",
+			"resolved": "https://registry.npmjs.org/side-channel-list/-/side-channel-list-1.0.0.tgz",
+			"integrity": "sha512-FCLHtRD/gnpCiCHEiJLOwdmFP+wzCmDEkc9y7NsYxeF4u7Btsn1ZuwgwJGxImImHicJArLP4R0yX4c2KCrMrTA==",
+			"license": "MIT",
+			"dependencies": {
+				"es-errors": "^1.3.0",
+				"object-inspect": "^1.13.3"
+			},
+			"engines": {
+				"node": ">= 0.4"
+			},
+			"funding": {
+				"url": "https://github.com/sponsors/ljharb"
+			}
+		},
+		"node_modules/side-channel-map": {
+			"version": "1.0.1",
+			"resolved": "https://registry.npmjs.org/side-channel-map/-/side-channel-map-1.0.1.tgz",
+			"integrity": "sha512-VCjCNfgMsby3tTdo02nbjtM/ewra6jPHmpThenkTYh8pG9ucZ/1P8So4u4FGBek/BjpOVsDCMoLA/iuBKIFXRA==",
+			"license": "MIT",
 			"dependencies": {
-				"call-bind": "^1.0.7",
+				"call-bound": "^1.0.2",
 				"es-errors": "^1.3.0",
-				"get-intrinsic": "^1.2.4",
-				"object-inspect": "^1.13.1"
+				"get-intrinsic": "^1.2.5",
+				"object-inspect": "^1.13.3"
+			},
+			"engines": {
+				"node": ">= 0.4"
+			},
+			"funding": {
+				"url": "https://github.com/sponsors/ljharb"
+			}
+		},
+		"node_modules/side-channel-weakmap": {
+			"version": "1.0.2",
+			"resolved": "https://registry.npmjs.org/side-channel-weakmap/-/side-channel-weakmap-1.0.2.tgz",
+			"integrity": "sha512-WPS/HvHQTYnHisLo9McqBHOJk2FkHO/tlpvldyrnem4aeQp4hai3gythswg6p01oSoTl58rcpiFAjF2br2Ak2A==",
+			"license": "MIT",
+			"dependencies": {
+				"call-bound": "^1.0.2",
+				"es-errors": "^1.3.0",
+				"get-intrinsic": "^1.2.5",
+				"object-inspect": "^1.13.3",
+				"side-channel-map": "^1.0.1"
 			},
 			"engines": {
 				"node": ">= 0.4"
@@ -15578,12 +16120,22 @@
 			"license": "MIT"
 		},
 		"node_modules/zod": {
-			"version": "3.23.8",
-			"resolved": "https://registry.npmjs.org/zod/-/zod-3.23.8.tgz",
-			"integrity": "sha512-XBx9AXhXktjUqnepgTiE5flcKIYWi/rme0Eaj+5Y0lftuGBq+jyRu/md4WnuxqgP1ubdpNCsYEYPxrzVHD8d6g==",
+			"version": "3.24.3",
+			"resolved": "https://registry.npmjs.org/zod/-/zod-3.24.3.tgz",
+			"integrity": "sha512-HhY1oqzWCQWuUqvBFnsyrtZRhyPeR7SUGv+C4+MsisMuVfSPx8HpwWqH8tRahSlt6M3PiFAcoeFhZAqIXTxoSg==",
+			"license": "MIT",
 			"funding": {
 				"url": "https://github.com/sponsors/colinhacks"
 			}
+		},
+		"node_modules/zod-to-json-schema": {
+			"version": "3.24.5",
+			"resolved": "https://registry.npmjs.org/zod-to-json-schema/-/zod-to-json-schema-3.24.5.tgz",
+			"integrity": "sha512-/AuWwMP+YqiPbsJx5D6TfgRTc4kTLjsh5SOcd4bLsfUg2RcEXrFMJl1DGgdHy2aCfsIA/cr/1JM0xcB2GZji8g==",
+			"license": "ISC",
+			"peerDependencies": {
+				"zod": "^3.24.1"
+			}
 		}
 	}
 }

package.json

@@ -47,6 +47,7 @@
 		"fs-extra": "^11.3.0",
 		"isomorphic-dompurify": "^2.13.0",
 		"js-yaml": "^4.1.0",
+		"jsonrepair": "^3.12.0",
 		"minimist": "^1.2.8",
 		"mongodb-memory-server": "^10.1.2",
 		"node-llama-cpp": "^3.6.0",
@@ -72,7 +73,9 @@
 		"@cliqz/adblocker-playwright": "^1.34.0",
 		"@gradio/client": "^1.8.0",
 		"@huggingface/hub": "^0.5.1",
-		"@huggingface/inference": "^2.8.1",
+		"@huggingface/inference": "^3.12.1",
+		"@huggingface/mcp-client": "^0.1.1",
+		"@huggingface/tasks": "^0.19.1",
 		"@huggingface/transformers": "^3.1.1",
 		"@iconify-json/bi": "^1.1.21",
 		"@playwright/browser-chromium": "^1.52.0",

src/lib/components/chat/ChatMessage.svelte

@@ -155,7 +155,7 @@
 		>
 			{#if message.files?.length}
 				<div class="flex h-fit flex-wrap gap-x-5 gap-y-2">
-					{#each message.files as file}
+					{#each message.files as file (file.value)}
 						<UploadedFile {file} canClose={false} />
 					{/each}
 				</div>
@@ -196,7 +196,9 @@
 				<div
 					class="prose max-w-none dark:prose-invert max-sm:prose-sm prose-headings:font-semibold prose-h1:text-lg prose-h2:text-base prose-h3:text-base prose-pre:bg-gray-800 dark:prose-pre:bg-gray-900"
 				>
-					<MarkdownRenderer content={message.content} sources={webSearchSources} />
+					{#key message.content}
+						<MarkdownRenderer content={message.content} sources={webSearchSources} />
+					{/key}
 				</div>
 			</div>
 

src/lib/components/chat/OpenReasoningResults.svelte

@@ -67,7 +67,9 @@
 	<div
 		class="space-y-4 border-t border-gray-200 px-5 pb-2 pt-2 text-sm text-gray-600 dark:border-gray-800 dark:text-gray-400"
 	>
-		<MarkdownRenderer {content} />
+		{#key content}
+			<MarkdownRenderer {content} />
+		{/key}
 	</div>
 </details>
 

src/lib/server/endpoints/aws/endpointAws.ts

@@ -64,7 +64,6 @@ export async function endpointAws(
 				inputs: prompt,
 			},
 			{
-				use_cache: false,
 				fetch: signedFetch,
 			}
 		);

src/lib/server/endpoints/endpoints.ts

@@ -31,6 +31,10 @@ import endpointLangserve, {
 import type { Tool, ToolCall, ToolResult } from "$lib/types/Tool";
 import type { ObjectId } from "mongodb";
 import { endpointLocal, endpointLocalParametersSchema } from "./local/endpointLocal";
+import {
+	endpointInferenceClient,
+	endpointInferenceClientParametersSchema,
+} from "./inference-client/endpointInferenceClient";
 
 export type EndpointMessage = Omit<Message, "id">;
 
@@ -77,6 +81,7 @@ export const endpoints = {
 	cohere: endpointCohere,
 	langserve: endpointLangserve,
 	local: endpointLocal,
+	inferenceClient: endpointInferenceClient,
 };
 
 export const endpointSchema = z.discriminatedUnion("type", [
@@ -94,5 +99,6 @@ export const endpointSchema = z.discriminatedUnion("type", [
 	endpointCohereParametersSchema,
 	endpointLangserveParametersSchema,
 	endpointLocalParametersSchema,
+	endpointInferenceClientParametersSchema,
 ]);
 export default endpoints;

src/lib/server/endpoints/inference-client/endpointInferenceClient.ts

@@ -0,0 +1,343 @@
+import { z } from "zod";
+import type { Endpoint, TextGenerationStreamOutputWithToolsAndWebSources } from "../endpoints";
+import { createImageProcessorOptionsValidator, makeImageProcessor } from "../images";
+import { INFERENCE_PROVIDERS, InferenceClient } from "@huggingface/inference";
+import { config } from "$lib/server/config";
+import type { Tool, ToolCall } from "$lib/types/Tool";
+import type { ChatCompletionStreamOutput } from "@huggingface/tasks";
+import type { FunctionDefinition } from "openai/resources/index.mjs";
+import type { ChatCompletionTool, FunctionParameters } from "openai/resources/index.mjs";
+import { logger } from "$lib/server/logger";
+import type { MessageFile } from "$lib/types/Message";
+import { v4 as uuidv4 } from "uuid";
+import type { Conversation } from "$lib/types/Conversation";
+import { downloadFile } from "$lib/server/files/downloadFile";
+import { jsonrepair } from "jsonrepair";
+type DeltaToolCall = NonNullable<
+	ChatCompletionStreamOutput["choices"][number]["delta"]["tool_calls"]
+>[number];
+
+function createChatCompletionToolsArray(tools: Tool[] | undefined): ChatCompletionTool[] {
+	const toolChoices = [] as ChatCompletionTool[];
+	if (tools === undefined) {
+		return toolChoices;
+	}
+
+	for (const t of tools) {
+		const requiredProperties = [] as string[];
+
+		const properties = {} as Record<string, unknown>;
+		for (const idx in t.inputs) {
+			const parameterDefinition = t.inputs[idx];
+
+			const parameter = {} as Record<string, unknown>;
+			switch (parameterDefinition.type) {
+				case "str":
+					parameter.type = "string";
+					break;
+				case "float":
+				case "int":
+					parameter.type = "number";
+					break;
+				case "bool":
+					parameter.type = "boolean";
+					break;
+				case "file":
+					throw new Error("File type's currently not supported");
+				default:
+					throw new Error(`Unknown tool IO type: ${t}`);
+			}
+
+			if ("description" in parameterDefinition) {
+				parameter.description = parameterDefinition.description;
+			}
+
+			if (parameterDefinition.paramType == "required") {
+				requiredProperties.push(t.inputs[idx].name);
+			}
+
+			properties[t.inputs[idx].name] = parameter;
+		}
+
+		const functionParameters: FunctionParameters = {
+			type: "object",
+			...(requiredProperties.length > 0 ? { required: requiredProperties } : {}),
+			properties,
+		};
+
+		const functionDefinition: FunctionDefinition = {
+			name: t.name,
+			description: t.description,
+			parameters: functionParameters,
+		};
+
+		const toolDefinition: ChatCompletionTool = {
+			type: "function",
+			function: functionDefinition,
+		};
+
+		toolChoices.push(toolDefinition);
+	}
+
+	return toolChoices;
+}
+
+export const endpointInferenceClientParametersSchema = z.object({
+	type: z.literal("inference-client"),
+	weight: z.number().int().positive().default(1),
+	model: z.any(),
+	provider: z.enum(INFERENCE_PROVIDERS).optional(),
+	supportedRoles: z
+		.array(z.enum(["user", "assistant", "system", "tool"]))
+		.default(["user", "assistant", "system", "tool"]),
+	modelName: z.string().optional(),
+	baseURL: z.string().optional(),
+	multimodal: z
+		.object({
+			image: createImageProcessorOptionsValidator({
+				supportedMimeTypes: [
+					"image/png",
+					"image/jpeg",
+					"image/webp",
+					"image/avif",
+					"image/tiff",
+					"image/gif",
+				],
+				preferredMimeType: "image/webp",
+				maxSizeInMB: Infinity,
+				maxWidth: 4096,
+				maxHeight: 4096,
+			}),
+		})
+		.default({}),
+});
+
+export async function endpointInferenceClient(
+	input: z.input<typeof endpointInferenceClientParametersSchema>
+): Promise<Endpoint> {
+	const { model, provider, modelName, baseURL, multimodal } =
+		endpointInferenceClientParametersSchema.parse(input);
+
+	if (!!provider && !!baseURL) {
+		throw new Error("provider and baseURL cannot both be provided");
+	}
+	const client = baseURL
+		? new InferenceClient(config.HF_TOKEN, { endpointUrl: baseURL })
+		: new InferenceClient(config.HF_TOKEN);
+
+	const imageProcessor = multimodal.image ? makeImageProcessor(multimodal.image) : undefined;
+
+	async function prepareFiles(files: MessageFile[], conversationId?: Conversation["_id"]) {
+		if (!imageProcessor) {
+			return [];
+		}
+		const processedFiles = await Promise.all(
+			files
+				.filter((file) => file.mime.startsWith("image/"))
+				.map(async (file) => {
+					if (file.type === "hash" && conversationId) {
+						file = await downloadFile(file.value, conversationId);
+					}
+					return imageProcessor(file);
+				})
+		);
+		return processedFiles.map((file) => ({
+			type: "image_url" as const,
+			image_url: {
+				url: `data:${file.mime};base64,${file.image.toString("base64")}`,
+			},
+		}));
+	}
+
+	return async ({ messages, generateSettings, tools, toolResults, preprompt, conversationId }) => {
+		/* eslint-disable @typescript-eslint/no-explicit-any */
+		let messagesArray = (await Promise.all(
+			messages.map(async (message) => {
+				return {
+					role: message.from,
+					content: [
+						...(await prepareFiles(message.files ?? [], conversationId)),
+						{ type: "text" as const, text: message.content },
+					],
+				};
+			})
+		)) as any[];
+
+		if (
+			!model.systemRoleSupported &&
+			messagesArray.length > 0 &&
+			messagesArray[0]?.role === "system"
+		) {
+			messagesArray[0].role = "user";
+		} else if (messagesArray[0].role !== "system") {
+			messagesArray.unshift({
+				role: "system",
+				content: preprompt ?? "",
+			});
+		}
+
+		if (toolResults && toolResults.length > 0) {
+			messagesArray = [
+				...messagesArray,
+				{
+					role: "assistant",
+					content: [
+						{
+							type: "text" as const,
+							text: "",
+						},
+					],
+					tool_calls: toolResults.map((toolResult) => ({
+						type: "function",
+						function: {
+							name: toolResult.call.name,
+							arguments: JSON.stringify(toolResult.call.parameters),
+						},
+						id: toolResult?.call?.toolId || uuidv4(),
+					})),
+				},
+				...toolResults.map((toolResult) => ({
+					role: model.systemRoleSupported ? "tool" : "user",
+					content: [
+						{
+							type: "text" as const,
+							text: JSON.stringify(toolResult),
+						},
+					],
+					tool_call_id: toolResult?.call?.toolId || uuidv4(),
+				})),
+			];
+		}
+
+		messagesArray = messagesArray.reduce((acc: typeof messagesArray, current) => {
+			if (acc.length === 0 || current.role !== acc[acc.length - 1].role) {
+				acc.push(current);
+			} else {
+				const prevMessage = acc[acc.length - 1];
+
+				prevMessage.content = [
+					...prevMessage.content.filter((item: any) => item.type !== "text"),
+					...current.content.filter((item: any) => item.type !== "text"),
+					{
+						type: "text" as const,
+						text: [
+							...prevMessage.content.filter((item: any) => item.type === "text"),
+							...current.content.filter((item: any) => item.type === "text"),
+						]
+							.map((item: any) => item.text)
+							.join("\n")
+							.replace(/^\n/, ""),
+					},
+				];
+
+				prevMessage.files = [...(prevMessage?.files ?? []), ...(current?.files ?? [])];
+
+				prevMessage.tool_calls = [
+					...(prevMessage?.tool_calls ?? []),
+					...(current?.tool_calls ?? []),
+				];
+			}
+			return acc;
+		}, []);
+		const toolCallChoices = createChatCompletionToolsArray(tools);
+		const stream = client.chatCompletionStream(
+			{
+				...model.parameters,
+				...generateSettings,
+				model: modelName ?? model.id ?? model.name,
+				provider: baseURL ? undefined : provider || ("hf-inference" as const),
+				messages: messagesArray,
+				...(toolCallChoices.length > 0 ? { tools: toolCallChoices, tool_choice: "auto" } : {}),
+				toolResults,
+			},
+			{
+				fetch: async (url, options) => {
+					return fetch(url, {
+						...options,
+						headers: {
+							...options?.headers,
+							"X-Use-Cache": "false",
+							"ChatUI-Conversation-ID": conversationId?.toString() ?? "",
+						},
+					});
+				},
+			}
+		);
+
+		let tokenId = 0;
+		let generated_text = "";
+		const finalToolCalls: DeltaToolCall[] = [];
+
+		async function* convertStream(): AsyncGenerator<
+			TextGenerationStreamOutputWithToolsAndWebSources,
+			void,
+			void
+		> {
+			for await (const chunk of stream) {
+				const token = chunk.choices?.[0]?.delta?.content || "";
+
+				generated_text += token;
+
+				const toolCalls = chunk.choices?.[0]?.delta?.tool_calls ?? [];
+
+				for (const toolCall of toolCalls) {
+					const index = toolCall.index ?? 0;
+
+					if (!finalToolCalls[index]) {
+						finalToolCalls[index] = toolCall;
+					} else {
+						if (finalToolCalls[index].function.arguments === undefined) {
+							finalToolCalls[index].function.arguments = "";
+						}
+						if (toolCall.function.arguments) {
+							finalToolCalls[index].function.arguments += toolCall.function.arguments;
+						}
+					}
+				}
+
+				yield {
+					token: {
+						id: tokenId++,
+						text: token,
+						logprob: 0,
+						special: false,
+					},
+					details: null,
+					generated_text: null,
+				};
+			}
+
+			let mappedToolCalls: ToolCall[] | undefined;
+			try {
+				if (finalToolCalls.length === 0) {
+					mappedToolCalls = undefined;
+				} else {
+					// Ensure finalToolCalls is an array
+					const toolCallsArray = Array.isArray(finalToolCalls) ? finalToolCalls : [finalToolCalls];
+
+					mappedToolCalls = toolCallsArray.map((tc) => ({
+						id: tc.id,
+						name: tc.function.name ?? "",
+						parameters: JSON.parse(jsonrepair(tc.function.arguments || "{}")),
+					}));
+				}
+			} catch (e) {
+				logger.error(e, "error mapping tool calls");
+			}
+
+			yield {
+				token: {
+					id: tokenId++,
+					text: "",
+					logprob: 0,
+					special: true,
+					toolCalls: mappedToolCalls,
+				},
+				generated_text,
+				details: null,
+			};
+		}
+
+		return convertStream();
+	};
+}

src/lib/server/endpoints/tgi/endpointTgi.ts

@@ -66,7 +66,6 @@ export function endpointTgi(input: z.input<typeof endpointTgiParametersSchema>):
 				accessToken,
 			},
 			{
-				use_cache: false,
 				fetch: async (endpointUrl, info) => {
 					if (info && authorization && !accessToken) {
 						// Set authorization header if it is defined and HF_TOKEN is empty
@@ -74,6 +73,7 @@ export function endpointTgi(input: z.input<typeof endpointTgiParametersSchema>):
 							...info.headers,
 							Authorization: authorization,
 							"ChatUI-Conversation-ID": conversationId?.toString() ?? "",
+							"X-Use-Cache": "false",
 						};
 					}
 					return fetch(endpointUrl, info);

src/lib/server/models.ts

@@ -12,7 +12,7 @@ import type { PreTrainedTokenizer } from "@huggingface/transformers";
 import JSON5 from "json5";
 import { getTokenizer } from "$lib/utils/getTokenizer";
 import { logger } from "$lib/server/logger";
-import { ToolResultStatus, type ToolInput } from "$lib/types/Tool";
+import { type ToolInput } from "$lib/types/Tool";
 import { join, dirname } from "path";
 import { resolveModelFile, readGgufFileInfo } from "node-llama-cpp";
 import { fileURLToPath } from "url";
@@ -195,13 +195,7 @@ async function getChatPromptRender(
 		process.exit();
 	}
 
-	const renderTemplate = ({
-		messages,
-		preprompt,
-		tools,
-		toolResults,
-		continueMessage,
-	}: ChatTemplateInput) => {
+	const renderTemplate = ({ messages, preprompt, tools, continueMessage }: ChatTemplateInput) => {
 		let formattedMessages: {
 			role: string;
 			content: string;
@@ -235,74 +229,6 @@ async function getChatPromptRender(
 			];
 		}
 
-		if (toolResults?.length) {
-			// todo: should update the command r+ tokenizer to support system messages at any location
-			// or use the `rag` mode without the citations
-			const id = m.id ?? m.name;
-
-			if (config.isHuggingChat && id.startsWith("CohereForAI")) {
-				formattedMessages = [
-					{
-						role: "user",
-						content:
-							"\n\n<results>\n" +
-							toolResults
-								.flatMap((result, idx) => {
-									if (result.status === ToolResultStatus.Error) {
-										return (
-											`Document: ${idx}\n` + `Tool "${result.call.name}" error\n` + result.message
-										);
-									}
-									return (
-										`Document: ${idx}\n` +
-										result.outputs
-											.flatMap((output) =>
-												Object.entries(output).map(([title, text]) => `${title}\n${text}`)
-											)
-											.join("\n")
-									);
-								})
-								.join("\n\n") +
-							"\n</results>",
-					},
-					...formattedMessages,
-				];
-			} else if (config.isHuggingChat && id.startsWith("meta-llama")) {
-				const results = toolResults.flatMap((result) => {
-					if (result.status === ToolResultStatus.Error) {
-						return [
-							{
-								tool_call_id: result.call.name,
-								output: "Error: " + result.message,
-							},
-						];
-					} else {
-						return result.outputs.map((output) => ({
-							tool_call_id: result.call.name,
-							output: JSON.stringify(output),
-						}));
-					}
-				});
-
-				formattedMessages = [
-					...formattedMessages,
-					{
-						role: "python",
-						content: JSON.stringify(results),
-					},
-				];
-			} else {
-				formattedMessages = [
-					...formattedMessages,
-					{
-						role: m.systemRoleSupported ? "system" : "user",
-						content: JSON.stringify(toolResults),
-					},
-				];
-			}
-			tools = [];
-		}
-
 		const mappedTools =
 			tools?.map((tool) => {
 				const inputs: Record<
@@ -380,6 +306,8 @@ const addEndpoint = (m: Awaited<ReturnType<typeof processModel>>) => ({
 						return endpoints.tgi(args);
 					case "local":
 						return endpoints.local(args);
+					case "inference-client":
+						return endpoints.inferenceClient(args);
 					case "anthropic":
 						return endpoints.anthropic(args);
 					case "anthropic-vertex":

src/lib/server/textGeneration/tools.ts

@@ -181,11 +181,9 @@ export async function* runTools(
 
 	// put fileMsg before last if files.length > 0
 	formattedMessages = files.length
-		? [...formattedMessages.slice(0, -1), fileMsg, ...formattedMessages.slice(-1)]
+		? [...formattedMessages.slice(0, 1), fileMsg, ...formattedMessages.slice(1)]
 		: messages;
 
-	let rawText = "";
-
 	const mappedTools = tools.map((tool) => ({
 		...tool,
 		inputs: tool.inputs.map((input) => ({
@@ -208,24 +206,6 @@ export async function* runTools(
 			continue;
 		}
 
-		if (output.token.text) {
-			rawText += output.token.text;
-		}
-
-		// if we dont see a tool call in the first 25 chars, something is going wrong and we abort
-		if (rawText.length > 100 && !(rawText.includes("```json") || rawText.includes("{"))) {
-			return [];
-		}
-
-		// if we see a directly_answer tool call, we skip the rest
-		if (
-			rawText.includes("directly_answer") ||
-			rawText.includes("directlyAnswer") ||
-			rawText.includes("directly-answer")
-		) {
-			return [];
-		}
-
 		// look for a code blocks of ```json and parse them
 		// if they're valid json, add them to the calls array
 		if (output.generated_text) {

src/lib/server/tools/web/search.ts

@@ -5,7 +5,7 @@ import { runWebSearch } from "../../websearch/runWebSearch";
 const websearch: ConfigTool = {
 	_id: new ObjectId("00000000000000000000000A"),
 	type: "config",
-	description: "Search the web for answers to the user's query",
+	description: "Search the web for up-to-date answers to the user's query.",
 	color: "blue",
 	icon: "wikis",
 	displayName: "Web Search",
@@ -16,7 +16,7 @@ const websearch: ConfigTool = {
 			name: "query",
 			type: "str",
 			description:
-				"A search query which will be used to fetch the most relevant snippets regarding the user's query",
+				"A search query which will be used to fetch the most relevant snippets regarding the user's query.",
 			paramType: "required",
 		},
 	],
@@ -35,7 +35,12 @@ const websearch: ConfigTool = {
 				{
 					websearch:
 						webSearchContext +
-						"\n\nWhen answering the question, you must reference the sources you used inline by wrapping the index in brackets like this: [1]. If multiple sources are used, you must reference each one of them without commas like this: [1][2][3].",
+						"\n\nWhen answering the question, you must reference the sources you used inline by wrapping the index in brackets like this: [1]. If multiple sources are used, you must reference each one of them without commas like this: [1][2][3]. This information was fetched live from the web, today on " +
+						new Date().toLocaleDateString("en-US", {
+							year: "numeric",
+							month: "long",
+							day: "numeric",
+						}),
 				},
 			],
 			display: false,

src/routes/models/+page.svelte

@@ -70,55 +70,62 @@
 								aria-hidden="true"
 							></div>
 						{/if}
-						{#if model.tools}
-							<span
-								title="This model supports tools."
-								class="ml-auto grid size-[21px] place-items-center rounded-lg border border-purple-300 dark:border-purple-700"
-								aria-label="Model supports tools"
-								role="img"
-							>
-								<CarbonTools class="text-xxs text-purple-700 dark:text-purple-500" />
-							</span>
-						{/if}
-						{#if model.multimodal}
-							<span
-								title="This model is multimodal and supports image inputs natively."
-								class="ml-auto flex size-[21px] items-center justify-center rounded-lg border border-blue-700 dark:border-blue-500"
-								aria-label="Model is multimodal"
-								role="img"
-							>
-								<CarbonImage class="text-xxs text-blue-700 dark:text-blue-500" />
-							</span>
-						{/if}
-						{#if model.reasoning}
-							<span
-								title="This model supports reasoning."
-								class="ml-auto grid size-[21px] place-items-center rounded-lg border border-purple-300 dark:border-purple-700"
-								aria-label="Model supports reasoning"
-								role="img"
-							>
-								<svg xmlns="http://www.w3.org/2000/svg" width="14" height="14" viewBox="0 0 32 32">
-									<path
-										class="stroke-purple-700"
-										style="stroke-width: 2; fill: none; stroke-linecap: round; stroke-linejoin: round; stroke-dasharray: 50;"
-										d="M16 6v3.33M16 6c0-2.65 3.25-4.3 5.4-2.62 1.2.95 1.6 2.65.95 4.04a3.63 3.63 0 0 1 4.61.16 3.45 3.45 0 0 1 .46 4.37 5.32 5.32 0 0 1 1.87 4.75c-.22 1.66-1.39 3.6-3.07 4.14M16 6c0-2.65-3.25-4.3-5.4-2.62a3.37 3.37 0 0 0-.95 4.04 3.65 3.65 0 0 0-4.6.16 3.37 3.37 0 0 0-.49 4.27 5.57 5.57 0 0 0-1.85 4.85 5.3 5.3 0 0 0 3.07 4.15M16 9.33v17.34m0-17.34c0 2.18 1.82 4 4 4m6.22 7.5c.67 1.3.56 2.91-.27 4.11a4.05 4.05 0 0 1-4.62 1.5c0 1.53-1.05 2.9-2.66 2.9A2.7 2.7 0 0 1 16 26.66m10.22-5.83a4.05 4.05 0 0 0-3.55-2.17m-16.9 2.18a4.05 4.05 0 0 0 .28 4.1c1 1.44 2.92 2.09 4.59 1.5 0 1.52 1.12 2.88 2.7 2.88A2.7 2.7 0 0 0 16 26.67M5.78 20.85a4.04 4.04 0 0 1 3.55-2.18"
-									/>
-								</svg>
-							</span>
-						{/if}
-						{#if model.id === $settings.activeModel}
-							<span
-								class="rounded-full border border-blue-500 bg-blue-500/5 px-2 py-0.5 text-xs text-blue-500 dark:border-blue-500 dark:bg-blue-500/10"
-							>
-								Active
-							</span>
-						{:else if index === 0}
-							<span
-								class="rounded-full border border-gray-300 px-2 py-0.5 text-xs text-gray-500 dark:border-gray-500 dark:text-gray-400"
-							>
-								Default
-							</span>
-						{/if}
+						<div class="flex items-center gap-1">
+							{#if model.tools}
+								<span
+									title="This model supports tools."
+									class="ml-auto grid size-[21px] place-items-center rounded-lg border border-purple-300 dark:border-purple-700"
+									aria-label="Model supports tools"
+									role="img"
+								>
+									<CarbonTools class="text-xxs text-purple-700 dark:text-purple-500" />
+								</span>
+							{/if}
+							{#if model.multimodal}
+								<span
+									title="This model is multimodal and supports image inputs natively."
+									class="ml-auto flex size-[21px] items-center justify-center rounded-lg border border-blue-700 dark:border-blue-500"
+									aria-label="Model is multimodal"
+									role="img"
+								>
+									<CarbonImage class="text-xxs text-blue-700 dark:text-blue-500" />
+								</span>
+							{/if}
+							{#if model.reasoning}
+								<span
+									title="This model supports reasoning."
+									class="ml-auto grid size-[21px] place-items-center rounded-lg border border-purple-300 dark:border-purple-700"
+									aria-label="Model supports reasoning"
+									role="img"
+								>
+									<svg
+										xmlns="http://www.w3.org/2000/svg"
+										width="14"
+										height="14"
+										viewBox="0 0 32 32"
+									>
+										<path
+											class="stroke-purple-700"
+											style="stroke-width: 2; fill: none; stroke-linecap: round; stroke-linejoin: round; stroke-dasharray: 50;"
+											d="M16 6v3.33M16 6c0-2.65 3.25-4.3 5.4-2.62 1.2.95 1.6 2.65.95 4.04a3.63 3.63 0 0 1 4.61.16 3.45 3.45 0 0 1 .46 4.37 5.32 5.32 0 0 1 1.87 4.75c-.22 1.66-1.39 3.6-3.07 4.14M16 6c0-2.65-3.25-4.3-5.4-2.62a3.37 3.37 0 0 0-.95 4.04 3.65 3.65 0 0 0-4.6.16 3.37 3.37 0 0 0-.49 4.27 5.57 5.57 0 0 0-1.85 4.85 5.3 5.3 0 0 0 3.07 4.15M16 9.33v17.34m0-17.34c0 2.18 1.82 4 4 4m6.22 7.5c.67 1.3.56 2.91-.27 4.11a4.05 4.05 0 0 1-4.62 1.5c0 1.53-1.05 2.9-2.66 2.9A2.7 2.7 0 0 1 16 26.66m10.22-5.83a4.05 4.05 0 0 0-3.55-2.17m-16.9 2.18a4.05 4.05 0 0 0 .28 4.1c1 1.44 2.92 2.09 4.59 1.5 0 1.52 1.12 2.88 2.7 2.88A2.7 2.7 0 0 0 16 26.67M5.78 20.85a4.04 4.04 0 0 1 3.55-2.18"
+										/>
+									</svg>
+								</span>
+							{/if}
+							{#if model.id === $settings.activeModel}
+								<span
+									class="rounded-full border border-blue-500 bg-blue-500/5 px-2 py-0.5 text-xs text-blue-500 dark:border-blue-500 dark:bg-blue-500/10"
+								>
+									Active
+								</span>
+							{:else if index === 0}
+								<span
+									class="rounded-full border border-gray-300 px-2 py-0.5 text-xs text-gray-500 dark:border-gray-500 dark:text-gray-400"
+								>
+									Default
+								</span>
+							{/if}
+						</div>
 					</div>
 					<span class="flex items-center gap-2 font-semibold">
 						{model.displayName}