Onboard NEAR AI as an inference provider

packages/inference/README.md

@@ -52,6 +52,7 @@ Currently, we support the following providers:
 - [Fireworks AI](https://fireworks.ai)
 - [HF Inference](https://huggingface.co/docs/inference-providers/providers/hf-inference)
 - [Hyperbolic](https://hyperbolic.xyz)
+- [NEAR AI](https://near.ai)
 - [Nebius](https://studio.nebius.ai)
 - [Novita](https://novita.ai)
 - [Nscale](https://nscale.com)
@@ -93,6 +94,7 @@ Only a subset of models are supported when requesting third-party providers. You
 - [Fireworks AI supported models](https://huggingface.co/api/partners/fireworks-ai/models)
 - [HF Inference supported models](https://huggingface.co/api/partners/hf-inference/models)
 - [Hyperbolic supported models](https://huggingface.co/api/partners/hyperbolic/models)
+- [NEAR AI supported models](https://huggingface.co/api/partners/near-ai/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)

packages/inference/src/lib/getProviderHelper.ts

@@ -10,6 +10,7 @@ import * as Groq from "../providers/groq.js";
 import * as HFInference from "../providers/hf-inference.js";
 import * as Hyperbolic from "../providers/hyperbolic.js";
 import * as Nebius from "../providers/nebius.js";
+import * as NearAI from "../providers/near-ai.js";
 import * as Novita from "../providers/novita.js";
 import * as Nscale from "../providers/nscale.js";
 import * as OpenAI from "../providers/openai.js";
@@ -136,6 +137,9 @@ export const PROVIDERS: Record<InferenceProvider, Partial<Record<InferenceTask,
 		"text-generation": new Nebius.NebiusTextGenerationTask(),
 		"feature-extraction": new Nebius.NebiusFeatureExtractionTask(),
 	},
+	"near-ai": {
+		conversational: new NearAI.NearAIConversationalTask(),
+	},
 	novita: {
 		conversational: new Novita.NovitaConversationalTask(),
 		"text-generation": new Novita.NovitaTextGenerationTask(),

packages/inference/src/providers/consts.ts

@@ -29,6 +29,7 @@ export const HARDCODED_MODEL_INFERENCE_MAPPING: Record<
 	groq: {},
 	"hf-inference": {},
 	hyperbolic: {},
+	"near-ai": {},
 	nebius: {},
 	novita: {},
 	nscale: {},

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

@@ -0,0 +1,25 @@
+/**
+ * See the registered mapping of HF model ID => NEAR AI model ID here:
+ *
+ * https://huggingface.co/api/partners/near-ai/models
+ *
+ * This is a publicly available mapping.
+ *
+ * If you want to try to run inference for a new model locally before it's registered on huggingface.co,
+ * you can add it to the dictionary "HARDCODED_MODEL_ID_MAPPING" in consts.ts, for dev purposes.
+ *
+ * - If you work at NEAR AI and want to update this mapping, please use the model mapping API we provide on huggingface.co
+ * - If you're a community member and want to add a new supported HF model to NEAR AI, please open an issue on the present repo
+ * and we will tag NEAR AI team members.
+ *
+ * Thanks!
+ */
+import { BaseConversationalTask } from "./providerHelper.js";
+
+const NEAR_AI_API_BASE_URL = "https://cloud-api.near.ai";
+
+export class NearAIConversationalTask extends BaseConversationalTask {
+	constructor() {
+		super("near-ai", NEAR_AI_API_BASE_URL);
+	}
+}

packages/inference/src/types.ts

@@ -56,6 +56,7 @@ export const INFERENCE_PROVIDERS = [
 	"groq",
 	"hf-inference",
 	"hyperbolic",
+	"near-ai",
 	"nebius",
 	"novita",
 	"nscale",
@@ -93,6 +94,7 @@ export const PROVIDERS_HUB_ORGS: Record<InferenceProvider, string> = {
 	groq: "groq",
 	"hf-inference": "hf-inference",
 	hyperbolic: "Hyperbolic",
+	"near-ai": "NEAR-AI",
 	nebius: "nebius",
 	novita: "novita",
 	nscale: "nscale",

packages/inference/test/InferenceClient.spec.ts

@@ -2578,6 +2578,59 @@ describe.skip("InferenceClient", () => {
 		TIMEOUT
 	);
 
+	describe.concurrent(
+		"NEAR AI",
+		() => {
+			const client = new InferenceClient(env.HF_NEAR_AI_KEY ?? "dummy");
+
+			HARDCODED_MODEL_INFERENCE_MAPPING["near-ai"] = {
+				"openai/gpt-oss-120b": {
+					provider: "near-ai",
+					hfModelId: "openai/gpt-oss-120b",
+					providerId: "openai/gpt-oss-120b",
+					status: "live",
+					task: "conversational",
+				},
+			};
+
+			it("chatCompletion", async () => {
+				const res = await client.chatCompletion({
+					model: "openai/gpt-oss-120b",
+					provider: "near-ai",
+					messages: [{ role: "user", content: "Complete this sentence with words, one plus one is equal " }],
+				});
+				if (res.choices && res.choices.length > 0) {
+					const completion = res.choices[0].message?.content;
+					expect(completion).toContain("two");
+				}
+			});
+
+			it("chatCompletion stream", async () => {
+				const stream = client.chatCompletionStream({
+					model: "openai/gpt-oss-120b",
+					provider: "near-ai",
+					messages: [{ role: "user", content: "Say this is a test" }],
+					stream: true,
+				}) as AsyncGenerator<ChatCompletionStreamOutput>;
+
+				let fullResponse = "";
+				for await (const chunk of stream) {
+					if (chunk.choices && chunk.choices.length > 0) {
+						const content = chunk.choices[0].delta?.content;
+						if (content) {
+							fullResponse += content;
+						}
+					}
+				}
+
+				// Verify we got a meaningful response
+				expect(fullResponse).toBeTruthy();
+				expect(fullResponse.length).toBeGreaterThan(0);
+			});
+		},
+		TIMEOUT
+	);
+
 	describe.concurrent(
 		"clarifai",
 		() => {