@neon/ai-sdk-provider 0.0.0 → 0.6.0

package/LICENSE.md

@@ -0,0 +1,178 @@
+# Apache License 2.0
+
+                                Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS

package/README.md

@@ -0,0 +1,103 @@
+# @neon/ai-sdk-provider
+
+Community [Vercel AI SDK](https://ai-sdk.dev) provider for the [Neon](https://neon.com) AI Gateway. Requires **AI SDK v6** (`ai@^6`).
+
+The Neon AI Gateway is **branch-scoped**: each Neon project branch gets its own gateway host, and a platform token authorizes requests for that branch. This provider routes each model to the best gateway endpoint (Anthropic → native Messages, OpenAI → native Responses incl. **Codex**, everything else → unified OpenAI-compatible MLflow endpoint), so a single `neon('claude-...')` call reaches the whole catalog.
+
+Model ids use the canonical Neon (unprefixed) form — `claude-sonnet-4-6`, `gpt-5`, `gemini-2-5-flash` — matching the [`neon` provider on models.dev](https://models.dev). The typed catalog mirrors that provider exactly (kept in sync by a scheduled drift check), plus a few extra gateway-served ids that models.dev doesn't list yet (e.g. Codex, Llama, Qwen). Any other id — including the legacy `databricks-` prefixed form (`databricks-claude-sonnet-4-6`) — is still accepted as a plain string, so existing code keeps working.
+
+## Install
+
+```bash
+npm install @neon/ai-sdk-provider ai@^6
+```
+
+> **Requirements:** Node.js >= 20.19.
+
+## Configuration
+
+The gateway URL is branch-scoped, so both values come from the Neon Console (your project → a branch → **AI Gateway** tab), or from `neon env pull` / `neon dev`:
+
+```bash
+NEON_AI_GATEWAY_BASE_URL="https://<branch-id>-api.ai.<region>.aws.neon.tech"
+NEON_AI_GATEWAY_TOKEN="nt_live_..."
+```
+
+## Usage
+
+```ts
+import { neon } from "@neon/ai-sdk-provider";
+import { generateText } from "ai";
+
+// Reads NEON_AI_GATEWAY_BASE_URL + NEON_AI_GATEWAY_TOKEN from the environment.
+const { text } = await generateText({
+  model: neon("claude-haiku-4-5"), // or 'gpt-5-3-codex', etc.
+  prompt: "Summarize Postgres for me.",
+});
+```
+
+Or configure explicitly with `createNeon`:
+
+```ts
+import { createNeon } from "@neon/ai-sdk-provider";
+
+const neon = createNeon({
+  baseURL: process.env.NEON_AI_GATEWAY_BASE_URL,
+  apiKey: process.env.NEON_AI_GATEWAY_TOKEN,
+});
+```
+
+## Routing
+
+| Model family | Endpoint | Why |
+| --- | --- | --- |
+| Anthropic (`claude-*`) | native Messages API | streaming structured output + native reasoning |
+| OpenAI (`gpt-*`, `*-codex`) | native Responses API | Codex (native-only), native reasoning, image-gen tool |
+| Everything else (Gemini, Llama, Qwen, gpt-oss, ...) | unified MLflow endpoint | broad coverage; Gemini is here because its native endpoint does not support streaming |
+
+Routing matches on the model id, so both the canonical (`gpt-5`) and the legacy `databricks-`-prefixed (`databricks-gpt-5`) forms route identically.
+
+## Capabilities
+
+Verified across Anthropic, OpenAI (incl. Codex), Google, and Meta models: `generateText` / `streamText` (text, system prompts, multi-turn), tool calling (single and multi-step, generate and stream), `generateObject` / `streamObject`, and image (vision) input.
+
+For MLflow-routed models, the provider detects the model family and drops parameters a backend rejects (e.g. penalties/`seed` for Llama, `reasoningEffort` for Gemini) with an AI SDK warning (`result.warnings`) instead of failing the request.
+
+## Image generation
+
+Available on OpenAI models via the Responses `image_generation` tool (there is no `generateImage()` image-model endpoint). Use `streamText` — streaming returns the image as a `tool-result` part and avoids the gateway's non-streaming response-size cap and read timeout:
+
+```ts
+import { streamText } from "ai";
+import { neon } from "@neon/ai-sdk-provider";
+
+const result = streamText({
+  model: neon("gpt-5-mini"),
+  prompt: "Generate an image of a red apple on a wooden table",
+  tools: { image: neon.tools.imageGeneration({ partialImages: 3 }) },
+});
+
+for await (const part of result.fullStream) {
+  if (part.type === "tool-result" && "result" in part.output) {
+    const png = Buffer.from(part.output.result as string, "base64");
+    // save or use the image
+  }
+}
+```
+
+## Limitations
+
+- `generateImage()` and embeddings (`embed` / `embedMany`) are not offered by the gateway and throw `NoSuchModelError`.
+- `gpt-oss-*` models return a non-standard ("harmony") response shape on the unified endpoint and are not fully supported.
+- OpenAI Responses multi-turn tool flows (`generateText` + `stepCountIs`) can return 502 from the gateway; tool calling is covered on Anthropic/Google/Meta in e2e.
+
+## End-to-end tests
+
+Against a live branch with AI Gateway enabled:
+
+```bash
+cp .env.example .env   # fill NEON_AI_GATEWAY_BASE_URL + NEON_AI_GATEWAY_TOKEN from `neon env pull`
+pnpm test:e2e
+```
+
+The matrix covers one models.dev `neon` model per family (Anthropic, OpenAI, Codex, Gemini, Meta) across `generateText`, `streamText`, `generateObject`, tool calling, and `neon.tools.imageGeneration`. Skipped when gateway env vars are absent.

package/dist/index.d.ts

@@ -0,0 +1,7 @@
+import { NeonAnthropicLanguageModel } from "./lib/neon-anthropic-language-model.js";
+import { NeonChatLanguageModel } from "./lib/neon-chat-language-model.js";
+import { NEON_EXTRA_MODEL_IDS, NEON_MODELS_DEV_IDS, NeonChatModelId, NeonKnownModelId } from "./lib/neon-chat-options.js";
+import { NeonModelCapabilities, NeonModelFamily, NeonModelRoute, getNeonModelCapabilities, getNeonModelRoute } from "./lib/neon-model-capabilities.js";
+import { NeonResponsesLanguageModel } from "./lib/neon-responses-language-model.js";
+import { NeonErrorData, NeonProvider, NeonProviderSettings, createNeon, neon } from "./lib/provider.js";
+export { NEON_EXTRA_MODEL_IDS, NEON_MODELS_DEV_IDS, NeonAnthropicLanguageModel, NeonChatLanguageModel, type NeonChatModelId, type NeonErrorData, type NeonKnownModelId, type NeonModelCapabilities, type NeonModelFamily, type NeonModelRoute, type NeonProvider, type NeonProviderSettings, NeonResponsesLanguageModel, createNeon, getNeonModelCapabilities, getNeonModelRoute, neon };

package/dist/index.js

@@ -0,0 +1,7 @@
+import { NeonAnthropicLanguageModel } from "./lib/neon-anthropic-language-model.js";
+import { getNeonModelCapabilities, getNeonModelRoute } from "./lib/neon-model-capabilities.js";
+import { NeonChatLanguageModel } from "./lib/neon-chat-language-model.js";
+import { NEON_EXTRA_MODEL_IDS, NEON_MODELS_DEV_IDS } from "./lib/neon-chat-options.js";
+import { NeonResponsesLanguageModel } from "./lib/neon-responses-language-model.js";
+import { createNeon, neon } from "./lib/provider.js";
+export { NEON_EXTRA_MODEL_IDS, NEON_MODELS_DEV_IDS, NeonAnthropicLanguageModel, NeonChatLanguageModel, NeonResponsesLanguageModel, createNeon, getNeonModelCapabilities, getNeonModelRoute, neon };

package/dist/lib/neon-anthropic-language-model.d.ts

@@ -0,0 +1,23 @@
+import { AnthropicMessagesLanguageModel } from "@ai-sdk/anthropic/internal";
+import { LanguageModelV3, LanguageModelV3CallOptions, LanguageModelV3GenerateResult, LanguageModelV3StreamResult } from "@ai-sdk/provider";
+
+//#region src/lib/neon-anthropic-language-model.d.ts
+
+/**
+ * Language model for Anthropic models served via the Neon AI Gateway's native
+ * Messages API (`/ai-gateway/anthropic/v1/messages`). Unlocks streaming
+ * structured output and native reasoning for Claude.
+ *
+ * The shared Anthropic model defaults to fine-grained tool-input streaming
+ * (`eager_input_streaming: true`) on streaming tool calls, which the gateway
+ * rejects (`Extra inputs are not permitted`). We disable it via the model's own
+ * `toolStreaming` option so streaming tool calls work. Users can still override.
+ */
+declare class NeonAnthropicLanguageModel extends AnthropicMessagesLanguageModel implements LanguageModelV3 {
+  private withGatewayCompat;
+  doGenerate(options: LanguageModelV3CallOptions): Promise<LanguageModelV3GenerateResult>;
+  doStream(options: LanguageModelV3CallOptions): Promise<LanguageModelV3StreamResult>;
+}
+//#endregion
+export { NeonAnthropicLanguageModel };
+//# sourceMappingURL=neon-anthropic-language-model.d.ts.map

package/dist/lib/neon-anthropic-language-model.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"neon-anthropic-language-model.d.ts","names":[],"sources":["../../src/lib/neon-anthropic-language-model.ts"],"mappings":";;;;;;;AAkBA;;;;;;;;AACS,cADI,0BAAA,SACJ,8BAAA,YACG,eADH,CAAA;UACG,iBAAA;EAAe,UAAA,CAAA,OAAA,EAoBhB,0BApBgB,CAAA,EAqBvB,OArBuB,CAqBf,6BArBe,CAAA;oBA0BhB,6BACP,QAAQ"}

package/dist/lib/neon-anthropic-language-model.js

@@ -0,0 +1,38 @@
+import { AnthropicMessagesLanguageModel } from "@ai-sdk/anthropic/internal";
+//#region src/lib/neon-anthropic-language-model.ts
+/**
+* Language model for Anthropic models served via the Neon AI Gateway's native
+* Messages API (`/ai-gateway/anthropic/v1/messages`). Unlocks streaming
+* structured output and native reasoning for Claude.
+*
+* The shared Anthropic model defaults to fine-grained tool-input streaming
+* (`eager_input_streaming: true`) on streaming tool calls, which the gateway
+* rejects (`Extra inputs are not permitted`). We disable it via the model's own
+* `toolStreaming` option so streaming tool calls work. Users can still override.
+*/
+var NeonAnthropicLanguageModel = class extends AnthropicMessagesLanguageModel {
+	withGatewayCompat(options) {
+		const anthropic = options.providerOptions?.anthropic;
+		if (anthropic != null && "toolStreaming" in anthropic) return options;
+		return {
+			...options,
+			providerOptions: {
+				...options.providerOptions,
+				anthropic: {
+					...anthropic,
+					toolStreaming: false
+				}
+			}
+		};
+	}
+	doGenerate(options) {
+		return super.doGenerate(this.withGatewayCompat(options));
+	}
+	doStream(options) {
+		return super.doStream(this.withGatewayCompat(options));
+	}
+};
+//#endregion
+export { NeonAnthropicLanguageModel };
+
+//# sourceMappingURL=neon-anthropic-language-model.js.map

package/dist/lib/neon-anthropic-language-model.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"neon-anthropic-language-model.js","names":[],"sources":["../../src/lib/neon-anthropic-language-model.ts"],"sourcesContent":["import { AnthropicMessagesLanguageModel } from \"@ai-sdk/anthropic/internal\";\nimport type {\n\tLanguageModelV3,\n\tLanguageModelV3CallOptions,\n\tLanguageModelV3GenerateResult,\n\tLanguageModelV3StreamResult,\n} from \"@ai-sdk/provider\";\n\n/**\n * Language model for Anthropic models served via the Neon AI Gateway's native\n * Messages API (`/ai-gateway/anthropic/v1/messages`). Unlocks streaming\n * structured output and native reasoning for Claude.\n *\n * The shared Anthropic model defaults to fine-grained tool-input streaming\n * (`eager_input_streaming: true`) on streaming tool calls, which the gateway\n * rejects (`Extra inputs are not permitted`). We disable it via the model's own\n * `toolStreaming` option so streaming tool calls work. Users can still override.\n */\nexport class NeonAnthropicLanguageModel\n\textends AnthropicMessagesLanguageModel\n\timplements LanguageModelV3\n{\n\tprivate withGatewayCompat(\n\t\toptions: LanguageModelV3CallOptions,\n\t): LanguageModelV3CallOptions {\n\t\tconst anthropic = options.providerOptions?.anthropic;\n\t\t// Respect an explicit user setting.\n\t\tif (anthropic != null && \"toolStreaming\" in anthropic) {\n\t\t\treturn options;\n\t\t}\n\t\treturn {\n\t\t\t...options,\n\t\t\tproviderOptions: {\n\t\t\t\t...options.providerOptions,\n\t\t\t\tanthropic: { ...anthropic, toolStreaming: false },\n\t\t\t},\n\t\t};\n\t}\n\n\toverride doGenerate(\n\t\toptions: LanguageModelV3CallOptions,\n\t): Promise<LanguageModelV3GenerateResult> {\n\t\treturn super.doGenerate(this.withGatewayCompat(options));\n\t}\n\n\toverride doStream(\n\t\toptions: LanguageModelV3CallOptions,\n\t): Promise<LanguageModelV3StreamResult> {\n\t\treturn super.doStream(this.withGatewayCompat(options));\n\t}\n}\n"],"mappings":";;;;;;;;;;;;AAkBA,IAAa,6BAAb,cACS,+BAET;CACC,kBACC,SAC6B;EAC7B,MAAM,YAAY,QAAQ,iBAAiB;EAE3C,IAAI,aAAa,QAAQ,mBAAmB,WAC3C,OAAO;EAER,OAAO;GACN,GAAG;GACH,iBAAiB;IAChB,GAAG,QAAQ;IACX,WAAW;KAAE,GAAG;KAAW,eAAe;IAAM;GACjD;EACD;CACD;CAEA,WACC,SACyC;EACzC,OAAO,MAAM,WAAW,KAAK,kBAAkB,OAAO,CAAC;CACxD;CAEA,SACC,SACuC;EACvC,OAAO,MAAM,SAAS,KAAK,kBAAkB,OAAO,CAAC;CACtD;AACD"}

package/dist/lib/neon-capabilities.d.ts

@@ -0,0 +1,20 @@
+import { LanguageModelV3CallOptions, LanguageModelV3StreamPart, SharedV3Warning } from "@ai-sdk/provider";
+
+//#region src/lib/neon-capabilities.d.ts
+
+/**
+ * Drop call options the resolved model's upstream backend does not accept and
+ * collect a warning for each, so callers get a clear signal instead of a hard
+ * `400` from the gateway.
+ */
+declare function applyNeonCapabilities(modelId: string, options: LanguageModelV3CallOptions): {
+  options: LanguageModelV3CallOptions;
+  warnings: SharedV3Warning[];
+};
+/**
+ * Merge additional warnings into the `stream-start` part of a model stream.
+ */
+declare function mergeStreamStartWarnings(extra: SharedV3Warning[]): TransformStream<LanguageModelV3StreamPart, LanguageModelV3StreamPart>;
+//#endregion
+export { applyNeonCapabilities, mergeStreamStartWarnings };
+//# sourceMappingURL=neon-capabilities.d.ts.map

package/dist/lib/neon-capabilities.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"neon-capabilities.d.ts","names":[],"sources":["../../src/lib/neon-capabilities.ts"],"mappings":";;;;;;AAaA;;;AAGc,iBAHE,qBAAA,CAGF,OAAA,EAAA,MAAA,EAAA,OAAA,EADJ,0BACI,CAAA,EAAA;SAAsC,EAAtC,0BAAsC;EAAe,QAAA,EAAf,eAAe,EAAA;AA4FnE,CAAA;;;;AAAiE,iBAAjD,wBAAA,CAAiD,KAAA,EAAjB,eAAiB,EAAA,CAAA,EAAA,eAAA,CAAA,yBAAA,EAAA,yBAAA,CAAA"}

package/dist/lib/neon-capabilities.js

@@ -0,0 +1,97 @@
+import { getNeonModelCapabilities } from "./neon-model-capabilities.js";
+//#region src/lib/neon-capabilities.ts
+/**
+* Drop call options the resolved model's upstream backend does not accept and
+* collect a warning for each, so callers get a clear signal instead of a hard
+* `400` from the gateway.
+*/
+function applyNeonCapabilities(modelId, options) {
+	const caps = getNeonModelCapabilities(modelId);
+	const warnings = [];
+	const patch = {};
+	const dropUnsupported = (feature) => {
+		warnings.push({
+			type: "unsupported",
+			feature,
+			details: `${feature} is not supported by the Neon AI Gateway for ${caps.family} model "${modelId}" and was dropped.`
+		});
+	};
+	if (options.temperature != null && !caps.supportsTemperature) {
+		patch.temperature = void 0;
+		dropUnsupported("temperature");
+	}
+	if (options.topP != null && !caps.supportsTopP) {
+		patch.topP = void 0;
+		dropUnsupported("topP");
+	}
+	const effectiveTemperature = "temperature" in patch ? patch.temperature : options.temperature;
+	const effectiveTopP = "topP" in patch ? patch.topP : options.topP;
+	if (caps.temperatureTopPMutuallyExclusive && effectiveTemperature != null && effectiveTopP != null) {
+		patch.topP = void 0;
+		warnings.push({
+			type: "compatibility",
+			feature: "topP",
+			details: `${caps.family} models accept only one of temperature or topP; topP was dropped.`
+		});
+	}
+	if (options.frequencyPenalty != null && !caps.supportsPenalties) {
+		patch.frequencyPenalty = void 0;
+		dropUnsupported("frequencyPenalty");
+	}
+	if (options.presencePenalty != null && !caps.supportsPenalties) {
+		patch.presencePenalty = void 0;
+		dropUnsupported("presencePenalty");
+	}
+	if (options.seed != null && !caps.supportsSeed) {
+		patch.seed = void 0;
+		dropUnsupported("seed");
+	}
+	if (options.stopSequences != null && !caps.supportsStopSequences) {
+		patch.stopSequences = void 0;
+		dropUnsupported("stopSequences");
+	}
+	if (!caps.supportsReasoningEffort) {
+		const { providerOptions } = options;
+		if (providerOptions != null && Object.values(providerOptions).some((group) => "reasoningEffort" in group && group.reasoningEffort != null)) {
+			dropUnsupported("reasoningEffort");
+			if (providerOptions != null) {
+				const cleaned = {};
+				for (const [key, group] of Object.entries(providerOptions)) if ("reasoningEffort" in group) {
+					const { reasoningEffort: _removed, ...rest } = group;
+					cleaned[key] = rest;
+				} else cleaned[key] = group;
+				patch.providerOptions = cleaned;
+			}
+		}
+	}
+	if (Object.keys(patch).length === 0) return {
+		options,
+		warnings
+	};
+	return {
+		options: {
+			...options,
+			...patch
+		},
+		warnings
+	};
+}
+/**
+* Merge additional warnings into the `stream-start` part of a model stream.
+*/
+function mergeStreamStartWarnings(extra) {
+	let merged = false;
+	return new TransformStream({ transform(part, controller) {
+		if (!merged && part.type === "stream-start") {
+			merged = true;
+			controller.enqueue({
+				...part,
+				warnings: [...extra, ...part.warnings]
+			});
+		} else controller.enqueue(part);
+	} });
+}
+//#endregion
+export { applyNeonCapabilities, mergeStreamStartWarnings };
+
+//# sourceMappingURL=neon-capabilities.js.map

package/dist/lib/neon-capabilities.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"neon-capabilities.js","names":[],"sources":["../../src/lib/neon-capabilities.ts"],"sourcesContent":["import type {\n\tLanguageModelV3CallOptions,\n\tLanguageModelV3StreamPart,\n\tSharedV3ProviderOptions,\n\tSharedV3Warning,\n} from \"@ai-sdk/provider\";\nimport { getNeonModelCapabilities } from \"./neon-model-capabilities.js\";\n\n/**\n * Drop call options the resolved model's upstream backend does not accept and\n * collect a warning for each, so callers get a clear signal instead of a hard\n * `400` from the gateway.\n */\nexport function applyNeonCapabilities(\n\tmodelId: string,\n\toptions: LanguageModelV3CallOptions,\n): { options: LanguageModelV3CallOptions; warnings: SharedV3Warning[] } {\n\tconst caps = getNeonModelCapabilities(modelId);\n\tconst warnings: SharedV3Warning[] = [];\n\tconst patch: Partial<LanguageModelV3CallOptions> = {};\n\n\tconst dropUnsupported = (feature: string) => {\n\t\twarnings.push({\n\t\t\ttype: \"unsupported\",\n\t\t\tfeature,\n\t\t\tdetails: `${feature} is not supported by the Neon AI Gateway for ${caps.family} model \"${modelId}\" and was dropped.`,\n\t\t});\n\t};\n\n\tif (options.temperature != null && !caps.supportsTemperature) {\n\t\tpatch.temperature = undefined;\n\t\tdropUnsupported(\"temperature\");\n\t}\n\tif (options.topP != null && !caps.supportsTopP) {\n\t\tpatch.topP = undefined;\n\t\tdropUnsupported(\"topP\");\n\t}\n\n\t// Anthropic-style models accept only one of temperature / topP.\n\tconst effectiveTemperature =\n\t\t\"temperature\" in patch ? patch.temperature : options.temperature;\n\tconst effectiveTopP = \"topP\" in patch ? patch.topP : options.topP;\n\tif (\n\t\tcaps.temperatureTopPMutuallyExclusive &&\n\t\teffectiveTemperature != null &&\n\t\teffectiveTopP != null\n\t) {\n\t\tpatch.topP = undefined;\n\t\twarnings.push({\n\t\t\ttype: \"compatibility\",\n\t\t\tfeature: \"topP\",\n\t\t\tdetails: `${caps.family} models accept only one of temperature or topP; topP was dropped.`,\n\t\t});\n\t}\n\n\tif (options.frequencyPenalty != null && !caps.supportsPenalties) {\n\t\tpatch.frequencyPenalty = undefined;\n\t\tdropUnsupported(\"frequencyPenalty\");\n\t}\n\tif (options.presencePenalty != null && !caps.supportsPenalties) {\n\t\tpatch.presencePenalty = undefined;\n\t\tdropUnsupported(\"presencePenalty\");\n\t}\n\tif (options.seed != null && !caps.supportsSeed) {\n\t\tpatch.seed = undefined;\n\t\tdropUnsupported(\"seed\");\n\t}\n\n\tif (options.stopSequences != null && !caps.supportsStopSequences) {\n\t\tpatch.stopSequences = undefined;\n\t\tdropUnsupported(\"stopSequences\");\n\t}\n\n\tif (!caps.supportsReasoningEffort) {\n\t\tconst { providerOptions } = options;\n\t\tconst hasProviderEffort =\n\t\t\tproviderOptions != null &&\n\t\t\tObject.values(providerOptions).some(\n\t\t\t\t(group) =>\n\t\t\t\t\t\"reasoningEffort\" in group && group.reasoningEffort != null,\n\t\t\t);\n\n\t\tif (hasProviderEffort) {\n\t\t\tdropUnsupported(\"reasoningEffort\");\n\t\t\tif (providerOptions != null) {\n\t\t\t\tconst cleaned: SharedV3ProviderOptions = {};\n\t\t\t\tfor (const [key, group] of Object.entries(providerOptions)) {\n\t\t\t\t\tif (\"reasoningEffort\" in group) {\n\t\t\t\t\t\tconst { reasoningEffort: _removed, ...rest } = group;\n\t\t\t\t\t\tcleaned[key] = rest;\n\t\t\t\t\t} else {\n\t\t\t\t\t\tcleaned[key] = group;\n\t\t\t\t\t}\n\t\t\t\t}\n\t\t\t\tpatch.providerOptions = cleaned;\n\t\t\t}\n\t\t}\n\t}\n\n\tif (Object.keys(patch).length === 0) {\n\t\treturn { options, warnings };\n\t}\n\treturn { options: { ...options, ...patch }, warnings };\n}\n\n/**\n * Merge additional warnings into the `stream-start` part of a model stream.\n */\nexport function mergeStreamStartWarnings(extra: SharedV3Warning[]) {\n\tlet merged = false;\n\treturn new TransformStream<\n\t\tLanguageModelV3StreamPart,\n\t\tLanguageModelV3StreamPart\n\t>({\n\t\ttransform(part, controller) {\n\t\t\tif (!merged && part.type === \"stream-start\") {\n\t\t\t\tmerged = true;\n\t\t\t\tcontroller.enqueue({\n\t\t\t\t\t...part,\n\t\t\t\t\twarnings: [...extra, ...part.warnings],\n\t\t\t\t});\n\t\t\t} else {\n\t\t\t\tcontroller.enqueue(part);\n\t\t\t}\n\t\t},\n\t});\n}\n"],"mappings":";;;;;;;AAaA,SAAgB,sBACf,SACA,SACuE;CACvE,MAAM,OAAO,yBAAyB,OAAO;CAC7C,MAAM,WAA8B,CAAC;CACrC,MAAM,QAA6C,CAAC;CAEpD,MAAM,mBAAmB,YAAoB;EAC5C,SAAS,KAAK;GACb,MAAM;GACN;GACA,SAAS,GAAG,QAAQ,+CAA+C,KAAK,OAAO,UAAU,QAAQ;EAClG,CAAC;CACF;CAEA,IAAI,QAAQ,eAAe,QAAQ,CAAC,KAAK,qBAAqB;EAC7D,MAAM,cAAc,KAAA;EACpB,gBAAgB,aAAa;CAC9B;CACA,IAAI,QAAQ,QAAQ,QAAQ,CAAC,KAAK,cAAc;EAC/C,MAAM,OAAO,KAAA;EACb,gBAAgB,MAAM;CACvB;CAGA,MAAM,uBACL,iBAAiB,QAAQ,MAAM,cAAc,QAAQ;CACtD,MAAM,gBAAgB,UAAU,QAAQ,MAAM,OAAO,QAAQ;CAC7D,IACC,KAAK,oCACL,wBAAwB,QACxB,iBAAiB,MAChB;EACD,MAAM,OAAO,KAAA;EACb,SAAS,KAAK;GACb,MAAM;GACN,SAAS;GACT,SAAS,GAAG,KAAK,OAAO;EACzB,CAAC;CACF;CAEA,IAAI,QAAQ,oBAAoB,QAAQ,CAAC,KAAK,mBAAmB;EAChE,MAAM,mBAAmB,KAAA;EACzB,gBAAgB,kBAAkB;CACnC;CACA,IAAI,QAAQ,mBAAmB,QAAQ,CAAC,KAAK,mBAAmB;EAC/D,MAAM,kBAAkB,KAAA;EACxB,gBAAgB,iBAAiB;CAClC;CACA,IAAI,QAAQ,QAAQ,QAAQ,CAAC,KAAK,cAAc;EAC/C,MAAM,OAAO,KAAA;EACb,gBAAgB,MAAM;CACvB;CAEA,IAAI,QAAQ,iBAAiB,QAAQ,CAAC,KAAK,uBAAuB;EACjE,MAAM,gBAAgB,KAAA;EACtB,gBAAgB,eAAe;CAChC;CAEA,IAAI,CAAC,KAAK,yBAAyB;EAClC,MAAM,EAAE,oBAAoB;EAQ5B,IANC,mBAAmB,QACnB,OAAO,OAAO,eAAe,CAAC,CAAC,MAC7B,UACA,qBAAqB,SAAS,MAAM,mBAAmB,IACzD,GAEsB;GACtB,gBAAgB,iBAAiB;GACjC,IAAI,mBAAmB,MAAM;IAC5B,MAAM,UAAmC,CAAC;IAC1C,KAAK,MAAM,CAAC,KAAK,UAAU,OAAO,QAAQ,eAAe,GACxD,IAAI,qBAAqB,OAAO;KAC/B,MAAM,EAAE,iBAAiB,UAAU,GAAG,SAAS;KAC/C,QAAQ,OAAO;IAChB,OACC,QAAQ,OAAO;IAGjB,MAAM,kBAAkB;GACzB;EACD;CACD;CAEA,IAAI,OAAO,KAAK,KAAK,CAAC,CAAC,WAAW,GACjC,OAAO;EAAE;EAAS;CAAS;CAE5B,OAAO;EAAE,SAAS;GAAE,GAAG;GAAS,GAAG;EAAM;EAAG;CAAS;AACtD;;;;AAKA,SAAgB,yBAAyB,OAA0B;CAClE,IAAI,SAAS;CACb,OAAO,IAAI,gBAGT,EACD,UAAU,MAAM,YAAY;EAC3B,IAAI,CAAC,UAAU,KAAK,SAAS,gBAAgB;GAC5C,SAAS;GACT,WAAW,QAAQ;IAClB,GAAG;IACH,UAAU,CAAC,GAAG,OAAO,GAAG,KAAK,QAAQ;GACtC,CAAC;EACF,OACC,WAAW,QAAQ,IAAI;CAEzB,EACD,CAAC;AACF"}

package/dist/lib/neon-chat-language-model.d.ts

@@ -0,0 +1,21 @@
+import { OpenAICompatibleChatLanguageModel } from "@ai-sdk/openai-compatible";
+import { LanguageModelV3, LanguageModelV3CallOptions, LanguageModelV3GenerateResult, LanguageModelV3StreamResult } from "@ai-sdk/provider";
+
+//#region src/lib/neon-chat-language-model.d.ts
+
+/**
+ * Language model for the Neon AI Gateway unified (MLflow) endpoint.
+ *
+ * Used for every model that is not routed to a provider-native endpoint
+ * (Gemini, Llama, Qwen, gpt-oss, ...). It is a thin specialization of the shared
+ * OpenAI-compatible chat model that adds per-model capability handling:
+ * parameters an upstream backend is known to reject are dropped and reported as
+ * warnings instead of failing the request.
+ */
+declare class NeonChatLanguageModel extends OpenAICompatibleChatLanguageModel implements LanguageModelV3 {
+  doGenerate(options: LanguageModelV3CallOptions): Promise<LanguageModelV3GenerateResult>;
+  doStream(options: LanguageModelV3CallOptions): Promise<LanguageModelV3StreamResult>;
+}
+//#endregion
+export { NeonChatLanguageModel };
+//# sourceMappingURL=neon-chat-language-model.d.ts.map

package/dist/lib/neon-chat-language-model.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"neon-chat-language-model.d.ts","names":[],"sources":["../../src/lib/neon-chat-language-model.ts"],"mappings":";;;;;;;AAqBA;;;;;;;AAmBI,cAnBS,qBAAA,SACJ,iCAAA,YACG,eAiBR,CAAA;YAlBK,CAAA,OAAA,EAIE,0BAJF,CAAA,EAKL,OALK,CAKG,6BALH,CAAA;UACG,CAAA,OAAA,EAgBD,0BAhBC,CAAA,EAiBR,OAjBQ,CAiBA,2BAjBA,CAAA;AAAe"}

package/dist/lib/neon-chat-language-model.js

@@ -0,0 +1,34 @@
+import { applyNeonCapabilities, mergeStreamStartWarnings } from "./neon-capabilities.js";
+import { OpenAICompatibleChatLanguageModel } from "@ai-sdk/openai-compatible";
+//#region src/lib/neon-chat-language-model.ts
+/**
+* Language model for the Neon AI Gateway unified (MLflow) endpoint.
+*
+* Used for every model that is not routed to a provider-native endpoint
+* (Gemini, Llama, Qwen, gpt-oss, ...). It is a thin specialization of the shared
+* OpenAI-compatible chat model that adds per-model capability handling:
+* parameters an upstream backend is known to reject are dropped and reported as
+* warnings instead of failing the request.
+*/
+var NeonChatLanguageModel = class extends OpenAICompatibleChatLanguageModel {
+	async doGenerate(options) {
+		const { options: adjusted, warnings } = applyNeonCapabilities(this.modelId, options);
+		const result = await super.doGenerate(adjusted);
+		return warnings.length > 0 ? {
+			...result,
+			warnings: [...warnings, ...result.warnings]
+		} : result;
+	}
+	async doStream(options) {
+		const { options: adjusted, warnings } = applyNeonCapabilities(this.modelId, options);
+		const result = await super.doStream(adjusted);
+		return warnings.length > 0 ? {
+			...result,
+			stream: result.stream.pipeThrough(mergeStreamStartWarnings(warnings))
+		} : result;
+	}
+};
+//#endregion
+export { NeonChatLanguageModel };
+
+//# sourceMappingURL=neon-chat-language-model.js.map

package/dist/lib/neon-chat-language-model.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"neon-chat-language-model.js","names":[],"sources":["../../src/lib/neon-chat-language-model.ts"],"sourcesContent":["import { OpenAICompatibleChatLanguageModel } from \"@ai-sdk/openai-compatible\";\nimport type {\n\tLanguageModelV3,\n\tLanguageModelV3CallOptions,\n\tLanguageModelV3GenerateResult,\n\tLanguageModelV3StreamResult,\n} from \"@ai-sdk/provider\";\nimport {\n\tapplyNeonCapabilities,\n\tmergeStreamStartWarnings,\n} from \"./neon-capabilities.js\";\n\n/**\n * Language model for the Neon AI Gateway unified (MLflow) endpoint.\n *\n * Used for every model that is not routed to a provider-native endpoint\n * (Gemini, Llama, Qwen, gpt-oss, ...). It is a thin specialization of the shared\n * OpenAI-compatible chat model that adds per-model capability handling:\n * parameters an upstream backend is known to reject are dropped and reported as\n * warnings instead of failing the request.\n */\nexport class NeonChatLanguageModel\n\textends OpenAICompatibleChatLanguageModel\n\timplements LanguageModelV3\n{\n\toverride async doGenerate(\n\t\toptions: LanguageModelV3CallOptions,\n\t): Promise<LanguageModelV3GenerateResult> {\n\t\tconst { options: adjusted, warnings } = applyNeonCapabilities(\n\t\t\tthis.modelId,\n\t\t\toptions,\n\t\t);\n\t\tconst result = await super.doGenerate(adjusted);\n\t\treturn warnings.length > 0\n\t\t\t? { ...result, warnings: [...warnings, ...result.warnings] }\n\t\t\t: result;\n\t}\n\n\toverride async doStream(\n\t\toptions: LanguageModelV3CallOptions,\n\t): Promise<LanguageModelV3StreamResult> {\n\t\tconst { options: adjusted, warnings } = applyNeonCapabilities(\n\t\t\tthis.modelId,\n\t\t\toptions,\n\t\t);\n\t\tconst result = await super.doStream(adjusted);\n\t\treturn warnings.length > 0\n\t\t\t? {\n\t\t\t\t\t...result,\n\t\t\t\t\tstream: result.stream.pipeThrough(\n\t\t\t\t\t\tmergeStreamStartWarnings(warnings),\n\t\t\t\t\t),\n\t\t\t\t}\n\t\t\t: result;\n\t}\n}\n"],"mappings":";;;;;;;;;;;;AAqBA,IAAa,wBAAb,cACS,kCAET;CACC,MAAe,WACd,SACyC;EACzC,MAAM,EAAE,SAAS,UAAU,aAAa,sBACvC,KAAK,SACL,OACD;EACA,MAAM,SAAS,MAAM,MAAM,WAAW,QAAQ;EAC9C,OAAO,SAAS,SAAS,IACtB;GAAE,GAAG;GAAQ,UAAU,CAAC,GAAG,UAAU,GAAG,OAAO,QAAQ;EAAE,IACzD;CACJ;CAEA,MAAe,SACd,SACuC;EACvC,MAAM,EAAE,SAAS,UAAU,aAAa,sBACvC,KAAK,SACL,OACD;EACA,MAAM,SAAS,MAAM,MAAM,SAAS,QAAQ;EAC5C,OAAO,SAAS,SAAS,IACtB;GACA,GAAG;GACH,QAAQ,OAAO,OAAO,YACrB,yBAAyB,QAAQ,CAClC;EACD,IACC;CACJ;AACD"}

package/dist/lib/neon-chat-options.d.ts

@@ -0,0 +1,21 @@
+//#region src/lib/neon-chat-options.d.ts
+/** The models.dev `neon` catalog (canonical, unprefixed ids). */
+declare const NEON_MODELS_DEV_IDS: readonly ["claude-opus-4-7", "claude-opus-4-6", "claude-opus-4-5", "claude-opus-4-1", "claude-sonnet-4-6", "claude-sonnet-4-5", "claude-sonnet-4", "claude-haiku-4-5", "gpt-5", "gpt-5-mini", "gpt-5-nano", "gpt-5-1", "gpt-5-2", "gpt-5-4", "gpt-5-4-mini", "gpt-5-4-nano", "gpt-5-5", "gpt-oss-120b", "gpt-oss-20b", "gemini-3-pro", "gemini-3-flash", "gemini-3-1-pro", "gemini-3-1-flash-lite", "gemini-2-5-pro", "gemini-2-5-flash"];
+/**
+ * Ids the gateway serves that the models.dev `neon` provider does not list.
+ * Verified to resolve against the gateway; retained for autocomplete. If
+ * models.dev later lists one of these, the drift check flags it for promotion
+ * into `NEON_MODELS_DEV_IDS`.
+ */
+declare const NEON_EXTRA_MODEL_IDS: readonly ["claude-opus-4-8", "gpt-5-2-codex", "gpt-5-3-codex", "gpt-5-5-pro", "gemini-3-5-flash", "gemma-3-12b", "llama-4-maverick", "meta-llama-3-3-70b-instruct", "meta-llama-3-1-8b-instruct", "qwen3-next-80b-a3b-instruct", "qwen35-122b-a10b"];
+/** A known Neon AI Gateway model id (models.dev catalog + gateway extras). */
+type NeonKnownModelId = (typeof NEON_MODELS_DEV_IDS)[number] | (typeof NEON_EXTRA_MODEL_IDS)[number];
+/**
+ * A Neon AI Gateway model id. Known ids are listed for autocomplete; any other
+ * id (e.g. a brand-new model, or the legacy `databricks-` prefixed form) is
+ * accepted via the `(string & {})` fallback.
+ */
+type NeonChatModelId = NeonKnownModelId | (string & {});
+//#endregion
+export { NEON_EXTRA_MODEL_IDS, NEON_MODELS_DEV_IDS, NeonChatModelId, NeonKnownModelId };
+//# sourceMappingURL=neon-chat-options.d.ts.map

package/dist/lib/neon-chat-options.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"neon-chat-options.d.ts","names":[],"sources":["../../src/lib/neon-chat-options.ts"],"mappings":";;AAiBa,cAAA,mBA8BH,EAAA,SAAA,CAAA,iBAAA,EAAA,iBAAA,EAAA,iBAAA,EAAA,iBAAA,EAAA,mBAAA,EAAA,mBAAA,EAAA,iBAAA,EAAA,kBAAA,EAAA,OAAA,EAAA,YAAA,EAAA,YAAA,EAAA,SAAA,EAAA,SAAA,EAAA,SAAA,EAAA,cAAA,EAAA,cAAA,EAAA,SAAA,EAAA,cAAA,EAAA,aAAA,EAAA,cAAA,EAAA,gBAAA,EAAA,gBAAA,EAAA,uBAAA,EAAA,gBAAA,EAAA,kBAAA,CAAA;AAQV;AAoBA;;;;AAE+B;AAOnB,cA7BC,oBA6BiB,EAAA,SAAgB,CAAA,iBAAA,EAAA,eAAA,EAAA,eAAA,EAAA,aAAA,EAAA,kBAAA,EAAA,aAAA,EAAA,kBAAA,EAAA,6BAAA,EAAA,4BAAA,EAAA,6BAAA,EAAA,kBAAA,CAAA;;KATlC,gBAAA,WACD,uCACA;;;;;;KAOC,eAAA,GAAkB"}

package/dist/lib/neon-chat-options.js

@@ -0,0 +1,52 @@
+//#region src/lib/neon-chat-options.ts
+/** The models.dev `neon` catalog (canonical, unprefixed ids). */
+const NEON_MODELS_DEV_IDS = [
+	"claude-opus-4-7",
+	"claude-opus-4-6",
+	"claude-opus-4-5",
+	"claude-opus-4-1",
+	"claude-sonnet-4-6",
+	"claude-sonnet-4-5",
+	"claude-sonnet-4",
+	"claude-haiku-4-5",
+	"gpt-5",
+	"gpt-5-mini",
+	"gpt-5-nano",
+	"gpt-5-1",
+	"gpt-5-2",
+	"gpt-5-4",
+	"gpt-5-4-mini",
+	"gpt-5-4-nano",
+	"gpt-5-5",
+	"gpt-oss-120b",
+	"gpt-oss-20b",
+	"gemini-3-pro",
+	"gemini-3-flash",
+	"gemini-3-1-pro",
+	"gemini-3-1-flash-lite",
+	"gemini-2-5-pro",
+	"gemini-2-5-flash"
+];
+/**
+* Ids the gateway serves that the models.dev `neon` provider does not list.
+* Verified to resolve against the gateway; retained for autocomplete. If
+* models.dev later lists one of these, the drift check flags it for promotion
+* into `NEON_MODELS_DEV_IDS`.
+*/
+const NEON_EXTRA_MODEL_IDS = [
+	"claude-opus-4-8",
+	"gpt-5-2-codex",
+	"gpt-5-3-codex",
+	"gpt-5-5-pro",
+	"gemini-3-5-flash",
+	"gemma-3-12b",
+	"llama-4-maverick",
+	"meta-llama-3-3-70b-instruct",
+	"meta-llama-3-1-8b-instruct",
+	"qwen3-next-80b-a3b-instruct",
+	"qwen35-122b-a10b"
+];
+//#endregion
+export { NEON_EXTRA_MODEL_IDS, NEON_MODELS_DEV_IDS };
+
+//# sourceMappingURL=neon-chat-options.js.map

package/dist/lib/neon-chat-options.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"neon-chat-options.js","names":[],"sources":["../../src/lib/neon-chat-options.ts"],"sourcesContent":["// Models served by the Neon AI Gateway. Anthropic and OpenAI models are routed\n// to their provider-native endpoints; everything else uses the unified MLflow\n// (OpenAI-compatible) endpoint.\n//\n// Ids use the canonical Neon (unprefixed) form, matching the `neon` provider on\n// models.dev. The gateway also accepts the legacy `databricks-` prefixed form\n// (the `databricks` provider on models.dev), which still resolves via the\n// `(string & {})` fallback on `NeonChatModelId`.\n//\n// `NEON_MODELS_DEV_IDS` mirrors the models.dev `neon` catalog exactly — the\n// `neon-catalog-drift.test.ts` maintainer check fails if the two diverge.\n// `NEON_EXTRA_MODEL_IDS` are additional ids the gateway serves that models.dev\n// does not (yet) list; they are kept for autocomplete and verified to resolve\n// against the gateway. The authoritative, always-current catalog is shown in\n// the Neon Console under the branch's \"AI Gateway\" tab.\n\n/** The models.dev `neon` catalog (canonical, unprefixed ids). */\nexport const NEON_MODELS_DEV_IDS = [\n\t// Anthropic (native Messages API)\n\t\"claude-opus-4-7\",\n\t\"claude-opus-4-6\",\n\t\"claude-opus-4-5\",\n\t\"claude-opus-4-1\",\n\t\"claude-sonnet-4-6\",\n\t\"claude-sonnet-4-5\",\n\t\"claude-sonnet-4\",\n\t\"claude-haiku-4-5\",\n\t// OpenAI (native Responses API)\n\t\"gpt-5\",\n\t\"gpt-5-mini\",\n\t\"gpt-5-nano\",\n\t\"gpt-5-1\",\n\t\"gpt-5-2\",\n\t\"gpt-5-4\",\n\t\"gpt-5-4-mini\",\n\t\"gpt-5-4-nano\",\n\t\"gpt-5-5\",\n\t// OpenAI open-weight (unified MLflow endpoint)\n\t\"gpt-oss-120b\",\n\t\"gpt-oss-20b\",\n\t// Google (unified MLflow endpoint)\n\t\"gemini-3-pro\",\n\t\"gemini-3-flash\",\n\t\"gemini-3-1-pro\",\n\t\"gemini-3-1-flash-lite\",\n\t\"gemini-2-5-pro\",\n\t\"gemini-2-5-flash\",\n] as const;\n\n/**\n * Ids the gateway serves that the models.dev `neon` provider does not list.\n * Verified to resolve against the gateway; retained for autocomplete. If\n * models.dev later lists one of these, the drift check flags it for promotion\n * into `NEON_MODELS_DEV_IDS`.\n */\nexport const NEON_EXTRA_MODEL_IDS = [\n\t// Anthropic (native Messages API)\n\t\"claude-opus-4-8\",\n\t// OpenAI (native Responses API) — Codex is served only natively\n\t\"gpt-5-2-codex\",\n\t\"gpt-5-3-codex\",\n\t\"gpt-5-5-pro\",\n\t// Google (unified MLflow endpoint)\n\t\"gemini-3-5-flash\",\n\t\"gemma-3-12b\",\n\t// Meta (unified MLflow endpoint)\n\t\"llama-4-maverick\",\n\t\"meta-llama-3-3-70b-instruct\",\n\t\"meta-llama-3-1-8b-instruct\",\n\t// Alibaba (unified MLflow endpoint)\n\t\"qwen3-next-80b-a3b-instruct\",\n\t\"qwen35-122b-a10b\",\n] as const;\n\n/** A known Neon AI Gateway model id (models.dev catalog + gateway extras). */\nexport type NeonKnownModelId =\n\t| (typeof NEON_MODELS_DEV_IDS)[number]\n\t| (typeof NEON_EXTRA_MODEL_IDS)[number];\n\n/**\n * A Neon AI Gateway model id. Known ids are listed for autocomplete; any other\n * id (e.g. a brand-new model, or the legacy `databricks-` prefixed form) is\n * accepted via the `(string & {})` fallback.\n */\nexport type NeonChatModelId = NeonKnownModelId | (string & {});\n"],"mappings":";;AAiBA,MAAa,sBAAsB;CAElC;CACA;CACA;CACA;CACA;CACA;CACA;CACA;CAEA;CACA;CACA;CACA;CACA;CACA;CACA;CACA;CACA;CAEA;CACA;CAEA;CACA;CACA;CACA;CACA;CACA;AACD;;;;;;;AAQA,MAAa,uBAAuB;CAEnC;CAEA;CACA;CACA;CAEA;CACA;CAEA;CACA;CACA;CAEA;CACA;AACD"}

package/dist/lib/neon-model-capabilities.d.ts

@@ -0,0 +1,41 @@
+//#region src/lib/neon-model-capabilities.d.ts
+type NeonModelFamily = "anthropic" | "google" | "openai" | "meta" | "other";
+/**
+ * Which gateway endpoint a model should be routed to.
+ *
+ * - `anthropic`: native Messages API — streaming structured output + native reasoning.
+ * - `openai`: native Responses API — models served only natively (e.g. Codex),
+ *   native reasoning, and the image-generation tool.
+ * - `mlflow`: the unified, OpenAI-compatible endpoint (the fallback). Gemini is
+ *   routed here because the gateway's native Gemini endpoint does not support
+ *   streaming (`streamGenerateContent` is rejected).
+ */
+type NeonModelRoute = "anthropic" | "openai" | "mlflow";
+declare function getNeonModelRoute(modelId: string): NeonModelRoute;
+interface NeonModelCapabilities {
+  family: NeonModelFamily;
+  supportsTemperature: boolean;
+  supportsTopP: boolean;
+  /**
+   * Whether `temperature` and `topP` may be sent together. Anthropic models
+   * accept only one of the two.
+   */
+  temperatureTopPMutuallyExclusive: boolean;
+  supportsPenalties: boolean;
+  supportsSeed: boolean;
+  supportsStopSequences: boolean;
+  supportsReasoningEffort: boolean;
+}
+/**
+ * Heuristic, prefix-based capability detection for MLflow-routed Neon models.
+ *
+ * The unified endpoint proxies to heterogeneous upstream providers, each of
+ * which accepts a different subset of the OpenAI-style parameters the AI SDK
+ * emits. Sending a parameter an upstream rejects results in a hard `400`, so we
+ * strip the parameters a family is known to reject and surface a warning
+ * instead. Unknown/untested models stay permissive (passed through unchanged).
+ */
+declare function getNeonModelCapabilities(modelId: string): NeonModelCapabilities;
+//#endregion
+export { NeonModelCapabilities, NeonModelFamily, NeonModelRoute, getNeonModelCapabilities, getNeonModelRoute };
+//# sourceMappingURL=neon-model-capabilities.d.ts.map

package/dist/lib/neon-model-capabilities.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"neon-model-capabilities.d.ts","names":[],"sources":["../../src/lib/neon-model-capabilities.ts"],"mappings":";KAAY,eAAA;AAAZ;AAiBA;AAEA;AAiBA;AAkCA;;;;;;KArDY,cAAA;iBAEI,iBAAA,mBAAoC;UAiBnC,qBAAA;UACR;;;;;;;;;;;;;;;;;;;;;;iBAiCO,wBAAA,mBAEb"}

package/dist/lib/neon-model-capabilities.js

@@ -0,0 +1,74 @@
+//#region src/lib/neon-model-capabilities.ts
+function getNeonModelRoute(modelId) {
+	const id = modelId.toLowerCase();
+	if (id.includes("claude")) return "anthropic";
+	if (id.includes("gpt-") && !id.includes("gpt-oss") || id.includes("codex")) return "openai";
+	return "mlflow";
+}
+const PERMISSIVE = {
+	supportsTemperature: true,
+	supportsTopP: true,
+	temperatureTopPMutuallyExclusive: false,
+	supportsPenalties: true,
+	supportsSeed: true,
+	supportsStopSequences: true,
+	supportsReasoningEffort: true
+};
+/**
+* Heuristic, prefix-based capability detection for MLflow-routed Neon models.
+*
+* The unified endpoint proxies to heterogeneous upstream providers, each of
+* which accepts a different subset of the OpenAI-style parameters the AI SDK
+* emits. Sending a parameter an upstream rejects results in a hard `400`, so we
+* strip the parameters a family is known to reject and surface a warning
+* instead. Unknown/untested models stay permissive (passed through unchanged).
+*/
+function getNeonModelCapabilities(modelId) {
+	const id = modelId.toLowerCase();
+	if (id.includes("claude")) return {
+		family: "anthropic",
+		supportsTemperature: true,
+		supportsTopP: true,
+		temperatureTopPMutuallyExclusive: true,
+		supportsPenalties: false,
+		supportsSeed: false,
+		supportsStopSequences: true,
+		supportsReasoningEffort: false
+	};
+	if (id.includes("gemini")) return {
+		family: "google",
+		...PERMISSIVE,
+		supportsReasoningEffort: false
+	};
+	if (/gpt-5/.test(id)) {
+		const hasMinorVersion = /gpt-5[.-]\d/.test(id);
+		return {
+			family: "openai",
+			supportsTemperature: hasMinorVersion,
+			supportsTopP: hasMinorVersion,
+			temperatureTopPMutuallyExclusive: false,
+			supportsPenalties: true,
+			supportsSeed: true,
+			supportsStopSequences: true,
+			supportsReasoningEffort: true
+		};
+	}
+	if (id.includes("llama")) return {
+		family: "meta",
+		supportsTemperature: true,
+		supportsTopP: true,
+		temperatureTopPMutuallyExclusive: false,
+		supportsPenalties: false,
+		supportsSeed: false,
+		supportsStopSequences: true,
+		supportsReasoningEffort: true
+	};
+	return {
+		family: "other",
+		...PERMISSIVE
+	};
+}
+//#endregion
+export { getNeonModelCapabilities, getNeonModelRoute };
+
+//# sourceMappingURL=neon-model-capabilities.js.map

package/dist/lib/neon-model-capabilities.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"neon-model-capabilities.js","names":[],"sources":["../../src/lib/neon-model-capabilities.ts"],"sourcesContent":["export type NeonModelFamily =\n\t| \"anthropic\"\n\t| \"google\"\n\t| \"openai\"\n\t| \"meta\"\n\t| \"other\";\n\n/**\n * Which gateway endpoint a model should be routed to.\n *\n * - `anthropic`: native Messages API — streaming structured output + native reasoning.\n * - `openai`: native Responses API — models served only natively (e.g. Codex),\n *   native reasoning, and the image-generation tool.\n * - `mlflow`: the unified, OpenAI-compatible endpoint (the fallback). Gemini is\n *   routed here because the gateway's native Gemini endpoint does not support\n *   streaming (`streamGenerateContent` is rejected).\n */\nexport type NeonModelRoute = \"anthropic\" | \"openai\" | \"mlflow\";\n\nexport function getNeonModelRoute(modelId: string): NeonModelRoute {\n\tconst id = modelId.toLowerCase();\n\tif (id.includes(\"claude\")) {\n\t\treturn \"anthropic\";\n\t}\n\t// OpenAI proprietary models (gpt-4/gpt-5 families, Codex) are served only via\n\t// the native Responses API. `gpt-oss` is open-weight and served on the\n\t// unified chat endpoint, so it is intentionally excluded here.\n\tif (\n\t\t(id.includes(\"gpt-\") && !id.includes(\"gpt-oss\")) ||\n\t\tid.includes(\"codex\")\n\t) {\n\t\treturn \"openai\";\n\t}\n\treturn \"mlflow\";\n}\n\nexport interface NeonModelCapabilities {\n\tfamily: NeonModelFamily;\n\tsupportsTemperature: boolean;\n\tsupportsTopP: boolean;\n\t/**\n\t * Whether `temperature` and `topP` may be sent together. Anthropic models\n\t * accept only one of the two.\n\t */\n\ttemperatureTopPMutuallyExclusive: boolean;\n\tsupportsPenalties: boolean;\n\tsupportsSeed: boolean;\n\tsupportsStopSequences: boolean;\n\tsupportsReasoningEffort: boolean;\n}\n\nconst PERMISSIVE: Omit<NeonModelCapabilities, \"family\"> = {\n\tsupportsTemperature: true,\n\tsupportsTopP: true,\n\ttemperatureTopPMutuallyExclusive: false,\n\tsupportsPenalties: true,\n\tsupportsSeed: true,\n\tsupportsStopSequences: true,\n\tsupportsReasoningEffort: true,\n};\n\n/**\n * Heuristic, prefix-based capability detection for MLflow-routed Neon models.\n *\n * The unified endpoint proxies to heterogeneous upstream providers, each of\n * which accepts a different subset of the OpenAI-style parameters the AI SDK\n * emits. Sending a parameter an upstream rejects results in a hard `400`, so we\n * strip the parameters a family is known to reject and surface a warning\n * instead. Unknown/untested models stay permissive (passed through unchanged).\n */\nexport function getNeonModelCapabilities(\n\tmodelId: string,\n): NeonModelCapabilities {\n\tconst id = modelId.toLowerCase();\n\n\t// Anthropic (Claude): rejects penalties and seed, accepts only one of\n\t// temperature/topP, and rejects the OpenAI `reasoning_effort` field.\n\tif (id.includes(\"claude\")) {\n\t\treturn {\n\t\t\tfamily: \"anthropic\",\n\t\t\tsupportsTemperature: true,\n\t\t\tsupportsTopP: true,\n\t\t\ttemperatureTopPMutuallyExclusive: true,\n\t\t\tsupportsPenalties: false,\n\t\t\tsupportsSeed: false,\n\t\t\tsupportsStopSequences: true,\n\t\t\tsupportsReasoningEffort: false,\n\t\t};\n\t}\n\n\t// Google (Gemini): accepts standard sampling params but rejects the OpenAI\n\t// `reasoning_effort` field (not part of Gemini's generation config).\n\tif (id.includes(\"gemini\")) {\n\t\treturn {\n\t\t\tfamily: \"google\",\n\t\t\t...PERMISSIVE,\n\t\t\tsupportsReasoningEffort: false,\n\t\t};\n\t}\n\n\t// OpenAI GPT-5 reasoning family (routed to the native Responses API). The\n\t// Responses model strips parameters the Responses API doesn't accept\n\t// (penalties, seed, stop), but its reasoning-model detection matches the bare\n\t// model id (`gpt-5`), which the gateway's optional `databricks-` prefix can\n\t// defeat. So we only handle the temperature/topP restriction here: the\n\t// original gpt-5 / gpt-5-mini / gpt-5-nano require the default temperature and\n\t// reject topP, while gpt-5.1+ (a minor version digit follows) accept them\n\t// again. The regex matches both prefixed and unprefixed ids.\n\tif (/gpt-5/.test(id)) {\n\t\tconst hasMinorVersion = /gpt-5[.-]\\d/.test(id);\n\t\treturn {\n\t\t\tfamily: \"openai\",\n\t\t\tsupportsTemperature: hasMinorVersion,\n\t\t\tsupportsTopP: hasMinorVersion,\n\t\t\ttemperatureTopPMutuallyExclusive: false,\n\t\t\tsupportsPenalties: true,\n\t\t\tsupportsSeed: true,\n\t\t\tsupportsStopSequences: true,\n\t\t\tsupportsReasoningEffort: true,\n\t\t};\n\t}\n\n\t// Meta (Llama): rejects penalties and seed; accepts sampling params and stop.\n\tif (id.includes(\"llama\")) {\n\t\treturn {\n\t\t\tfamily: \"meta\",\n\t\t\tsupportsTemperature: true,\n\t\t\tsupportsTopP: true,\n\t\t\ttemperatureTopPMutuallyExclusive: false,\n\t\t\tsupportsPenalties: false,\n\t\t\tsupportsSeed: false,\n\t\t\tsupportsStopSequences: true,\n\t\t\tsupportsReasoningEffort: true,\n\t\t};\n\t}\n\n\treturn { family: \"other\", ...PERMISSIVE };\n}\n"],"mappings":";AAmBA,SAAgB,kBAAkB,SAAiC;CAClE,MAAM,KAAK,QAAQ,YAAY;CAC/B,IAAI,GAAG,SAAS,QAAQ,GACvB,OAAO;CAKR,IACE,GAAG,SAAS,MAAM,KAAK,CAAC,GAAG,SAAS,SAAS,KAC9C,GAAG,SAAS,OAAO,GAEnB,OAAO;CAER,OAAO;AACR;AAiBA,MAAM,aAAoD;CACzD,qBAAqB;CACrB,cAAc;CACd,kCAAkC;CAClC,mBAAmB;CACnB,cAAc;CACd,uBAAuB;CACvB,yBAAyB;AAC1B;;;;;;;;;;AAWA,SAAgB,yBACf,SACwB;CACxB,MAAM,KAAK,QAAQ,YAAY;CAI/B,IAAI,GAAG,SAAS,QAAQ,GACvB,OAAO;EACN,QAAQ;EACR,qBAAqB;EACrB,cAAc;EACd,kCAAkC;EAClC,mBAAmB;EACnB,cAAc;EACd,uBAAuB;EACvB,yBAAyB;CAC1B;CAKD,IAAI,GAAG,SAAS,QAAQ,GACvB,OAAO;EACN,QAAQ;EACR,GAAG;EACH,yBAAyB;CAC1B;CAWD,IAAI,QAAQ,KAAK,EAAE,GAAG;EACrB,MAAM,kBAAkB,cAAc,KAAK,EAAE;EAC7C,OAAO;GACN,QAAQ;GACR,qBAAqB;GACrB,cAAc;GACd,kCAAkC;GAClC,mBAAmB;GACnB,cAAc;GACd,uBAAuB;GACvB,yBAAyB;EAC1B;CACD;CAGA,IAAI,GAAG,SAAS,OAAO,GACtB,OAAO;EACN,QAAQ;EACR,qBAAqB;EACrB,cAAc;EACd,kCAAkC;EAClC,mBAAmB;EACnB,cAAc;EACd,uBAAuB;EACvB,yBAAyB;CAC1B;CAGD,OAAO;EAAE,QAAQ;EAAS,GAAG;CAAW;AACzC"}

package/dist/lib/neon-responses-language-model.d.ts

@@ -0,0 +1,27 @@
+import { OpenAIResponsesLanguageModel } from "@ai-sdk/openai/internal";
+import { LanguageModelV3, LanguageModelV3CallOptions, LanguageModelV3GenerateResult, LanguageModelV3StreamResult } from "@ai-sdk/provider";
+
+//#region src/lib/neon-responses-language-model.d.ts
+
+/**
+ * Language model for OpenAI models served via the Neon AI Gateway's native
+ * Responses API (`/ai-gateway/openai/v1/responses`). This route is required for
+ * models that are only served natively (e.g. Codex) and unlocks native
+ * reasoning and the image-generation tool.
+ *
+ * The shared OpenAI Responses model shapes a request based on whether the model
+ * is a reasoning model, but its detection is brittle here: the gateway also
+ * accepts the legacy `databricks-` prefix, which defeats the upstream bare-id
+ * (`gpt-5`) match. For the GPT-5 reasoning family we set the model's own
+ * `forceReasoning` provider option so it applies the correct reasoning behavior
+ * for both id forms. Users can still override it.
+ */
+declare class NeonResponsesLanguageModel extends OpenAIResponsesLanguageModel implements LanguageModelV3 {
+  private get isReasoningFamily();
+  private withForcedReasoning;
+  doGenerate(options: LanguageModelV3CallOptions): Promise<LanguageModelV3GenerateResult>;
+  doStream(options: LanguageModelV3CallOptions): Promise<LanguageModelV3StreamResult>;
+}
+//#endregion
+export { NeonResponsesLanguageModel };
+//# sourceMappingURL=neon-responses-language-model.d.ts.map

package/dist/lib/neon-responses-language-model.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"neon-responses-language-model.d.ts","names":[],"sources":["../../src/lib/neon-responses-language-model.ts"],"mappings":";;;;;;;AAqBA;;;;;;;;;;AAE2B;cAFd,0BAAA,SACJ,4BAAA,YACG;;;sBA2BD,6BACP,QAAQ;oBAKD,6BACP,QAAQ"}

package/dist/lib/neon-responses-language-model.js

@@ -0,0 +1,45 @@
+import { OpenAIResponsesLanguageModel } from "@ai-sdk/openai/internal";
+//#region src/lib/neon-responses-language-model.ts
+/**
+* Language model for OpenAI models served via the Neon AI Gateway's native
+* Responses API (`/ai-gateway/openai/v1/responses`). This route is required for
+* models that are only served natively (e.g. Codex) and unlocks native
+* reasoning and the image-generation tool.
+*
+* The shared OpenAI Responses model shapes a request based on whether the model
+* is a reasoning model, but its detection is brittle here: the gateway also
+* accepts the legacy `databricks-` prefix, which defeats the upstream bare-id
+* (`gpt-5`) match. For the GPT-5 reasoning family we set the model's own
+* `forceReasoning` provider option so it applies the correct reasoning behavior
+* for both id forms. Users can still override it.
+*/
+var NeonResponsesLanguageModel = class extends OpenAIResponsesLanguageModel {
+	get isReasoningFamily() {
+		return /gpt-5/.test(this.modelId.toLowerCase());
+	}
+	withForcedReasoning(options) {
+		if (!this.isReasoningFamily) return options;
+		const openai = options.providerOptions?.openai;
+		if (openai != null && "forceReasoning" in openai) return options;
+		return {
+			...options,
+			providerOptions: {
+				...options.providerOptions,
+				openai: {
+					...openai,
+					forceReasoning: true
+				}
+			}
+		};
+	}
+	doGenerate(options) {
+		return super.doGenerate(this.withForcedReasoning(options));
+	}
+	doStream(options) {
+		return super.doStream(this.withForcedReasoning(options));
+	}
+};
+//#endregion
+export { NeonResponsesLanguageModel };
+
+//# sourceMappingURL=neon-responses-language-model.js.map

package/dist/lib/neon-responses-language-model.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"neon-responses-language-model.js","names":[],"sources":["../../src/lib/neon-responses-language-model.ts"],"sourcesContent":["import { OpenAIResponsesLanguageModel } from \"@ai-sdk/openai/internal\";\nimport type {\n\tLanguageModelV3,\n\tLanguageModelV3CallOptions,\n\tLanguageModelV3GenerateResult,\n\tLanguageModelV3StreamResult,\n} from \"@ai-sdk/provider\";\n\n/**\n * Language model for OpenAI models served via the Neon AI Gateway's native\n * Responses API (`/ai-gateway/openai/v1/responses`). This route is required for\n * models that are only served natively (e.g. Codex) and unlocks native\n * reasoning and the image-generation tool.\n *\n * The shared OpenAI Responses model shapes a request based on whether the model\n * is a reasoning model, but its detection is brittle here: the gateway also\n * accepts the legacy `databricks-` prefix, which defeats the upstream bare-id\n * (`gpt-5`) match. For the GPT-5 reasoning family we set the model's own\n * `forceReasoning` provider option so it applies the correct reasoning behavior\n * for both id forms. Users can still override it.\n */\nexport class NeonResponsesLanguageModel\n\textends OpenAIResponsesLanguageModel\n\timplements LanguageModelV3\n{\n\tprivate get isReasoningFamily(): boolean {\n\t\treturn /gpt-5/.test(this.modelId.toLowerCase());\n\t}\n\n\tprivate withForcedReasoning(\n\t\toptions: LanguageModelV3CallOptions,\n\t): LanguageModelV3CallOptions {\n\t\tif (!this.isReasoningFamily) {\n\t\t\treturn options;\n\t\t}\n\t\tconst openai = options.providerOptions?.openai;\n\t\t// Respect an explicit user setting.\n\t\tif (openai != null && \"forceReasoning\" in openai) {\n\t\t\treturn options;\n\t\t}\n\t\treturn {\n\t\t\t...options,\n\t\t\tproviderOptions: {\n\t\t\t\t...options.providerOptions,\n\t\t\t\topenai: { ...openai, forceReasoning: true },\n\t\t\t},\n\t\t};\n\t}\n\n\toverride doGenerate(\n\t\toptions: LanguageModelV3CallOptions,\n\t): Promise<LanguageModelV3GenerateResult> {\n\t\treturn super.doGenerate(this.withForcedReasoning(options));\n\t}\n\n\toverride doStream(\n\t\toptions: LanguageModelV3CallOptions,\n\t): Promise<LanguageModelV3StreamResult> {\n\t\treturn super.doStream(this.withForcedReasoning(options));\n\t}\n}\n"],"mappings":";;;;;;;;;;;;;;;AAqBA,IAAa,6BAAb,cACS,6BAET;CACC,IAAY,oBAA6B;EACxC,OAAO,QAAQ,KAAK,KAAK,QAAQ,YAAY,CAAC;CAC/C;CAEA,oBACC,SAC6B;EAC7B,IAAI,CAAC,KAAK,mBACT,OAAO;EAER,MAAM,SAAS,QAAQ,iBAAiB;EAExC,IAAI,UAAU,QAAQ,oBAAoB,QACzC,OAAO;EAER,OAAO;GACN,GAAG;GACH,iBAAiB;IAChB,GAAG,QAAQ;IACX,QAAQ;KAAE,GAAG;KAAQ,gBAAgB;IAAK;GAC3C;EACD;CACD;CAEA,WACC,SACyC;EACzC,OAAO,MAAM,WAAW,KAAK,oBAAoB,OAAO,CAAC;CAC1D;CAEA,SACC,SACuC;EACvC,OAAO,MAAM,SAAS,KAAK,oBAAoB,OAAO,CAAC;CACxD;AACD"}

package/dist/lib/provider.d.ts

@@ -0,0 +1,52 @@
+import { NeonChatModelId } from "./neon-chat-options.js";
+import { openai } from "@ai-sdk/openai";
+import { LanguageModelV3, ProviderV3 } from "@ai-sdk/provider";
+import { FetchFunction } from "@ai-sdk/provider-utils";
+import { z } from "zod/v4";
+
+//#region src/lib/provider.d.ts
+
+declare const neonErrorSchema: z.ZodObject<{
+  error: z.ZodObject<{
+    message: z.ZodString;
+    type: z.ZodOptional<z.ZodNullable<z.ZodString>>;
+    param: z.ZodOptional<z.ZodNullable<z.ZodUnknown>>;
+    code: z.ZodOptional<z.ZodNullable<z.ZodUnknown>>;
+  }, z.core.$strip>;
+}, z.core.$strip>;
+type NeonErrorData = z.infer<typeof neonErrorSchema>;
+interface NeonProviderSettings {
+  /**
+   * Neon AI Gateway base URL — the branch-scoped host root, e.g.
+   * `https://<branch-id>-api.ai.<region>.aws.neon.tech`. The gateway paths are
+   * appended internally. Falls back to the `NEON_AI_GATEWAY_BASE_URL` env var.
+   */
+  baseURL?: string;
+  /**
+   * Neon AI Gateway platform token (the `nt_live_...` value). Falls back to the
+   * `NEON_AI_GATEWAY_TOKEN` env var.
+   */
+  apiKey?: string;
+  /** Custom headers to include in the requests. */
+  headers?: Record<string, string>;
+  /** Custom fetch implementation (e.g. for testing or middleware). */
+  fetch?: FetchFunction;
+}
+interface NeonProvider extends ProviderV3 {
+  /** Creates a Neon AI Gateway model for text generation. */
+  (modelId: NeonChatModelId): LanguageModelV3;
+  /** Creates a Neon AI Gateway model for text generation. */
+  languageModel(modelId: NeonChatModelId): LanguageModelV3;
+  /** Creates a Neon AI Gateway chat model for text generation. */
+  chat(modelId: NeonChatModelId): LanguageModelV3;
+  /** OpenAI Responses tools (e.g. `imageGeneration`) for OpenAI-routed models. */
+  tools: typeof openai.tools;
+  /** @deprecated Use `embeddingModel` instead. */
+  textEmbeddingModel(modelId: string): never;
+}
+declare function createNeon(options?: NeonProviderSettings): NeonProvider;
+/** Default Neon provider instance (reads `NEON_AI_GATEWAY_*` env vars). */
+declare const neon: NeonProvider;
+//#endregion
+export { NeonErrorData, NeonProvider, NeonProviderSettings, createNeon, neon };
+//# sourceMappingURL=provider.d.ts.map

package/dist/lib/provider.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"provider.d.ts","names":[],"sources":["../../src/lib/provider.ts"],"mappings":";;;;;;;;cAmCM,iBAAe,CAAA,CAAA;;;;;IAAA,IAAA,eAAA,cAAA,aAAA,CAAA,CAAA;EAAA,CAAA,eAAA,CAAA;AASrB,CAAA,eAAY,CAAA;AAAa,KAAb,aAAA,GAAgB,CAAA,CAAE,KAAL,CAAA,OAAkB,eAAlB,CAAA;AAAkB,UA4C1B,oBAAA,CA5C0B;;AAAR;AA4CnC;;;SAkBS,CAAA,EAAA,MAAA;EAAa;AAGtB;;;QAE6B,CAAA,EAAA,MAAA;;SAGa,CAAA,EAX/B,MAW+B,CAAA,MAAA,EAAA,MAAA,CAAA;;OAGT,CAAA,EAXxB,aAWwB;;AARK,UAArB,YAAA,SAAqB,UAAA,CAAA;EAAU;EAiBhC,CAAA,OAAA,EAfL,eAee,CAAA,EAfG,eAeH;EAAA;eAAU,CAAA,OAAA,EAZZ,eAYY,CAAA,EAZM,eAYN;;EAAwC,IAAA,CAAA,OAAA,EAT7D,eAS6D,CAAA,EAT3C,eAS2C;EA0F/D;gBAhGE,MAAA,CAAe;;;;iBAMd,UAAA,WAAoB,uBAA4B;;cA0FnD,MAAI"}

package/dist/lib/provider.js

@@ -0,0 +1,126 @@
+import { NeonAnthropicLanguageModel } from "./neon-anthropic-language-model.js";
+import { getNeonModelRoute } from "./neon-model-capabilities.js";
+import { NeonChatLanguageModel } from "./neon-chat-language-model.js";
+import { NeonResponsesLanguageModel } from "./neon-responses-language-model.js";
+import { VERSION } from "./version.js";
+import { openai } from "@ai-sdk/openai";
+import { NoSuchModelError } from "@ai-sdk/provider";
+import { generateId, loadApiKey, loadSetting, withUserAgentSuffix, withoutTrailingSlash } from "@ai-sdk/provider-utils";
+import { z } from "zod/v4";
+//#region src/lib/provider.ts
+/**
+* Community Vercel AI SDK provider for the Neon AI Gateway.
+*
+* `createNeon()` returns a provider that routes each model to the best gateway
+* endpoint based on its id (Anthropic → native Messages, OpenAI → native
+* Responses incl. Codex, everything else → unified MLflow), so a single
+* `neon('claude-...')` call (same base URL + token) reaches the whole catalog.
+* Ids use the canonical Neon (unprefixed) form; the legacy `databricks-` prefix
+* is also accepted. Configure with the branch-scoped `NEON_AI_GATEWAY_BASE_URL` +
+* `NEON_AI_GATEWAY_TOKEN` emitted by `neonctl env pull` / `neon dev`, or pass
+* `baseURL` / `apiKey` explicitly.
+*/
+const neonErrorStructure = {
+	errorSchema: z.object({ error: z.object({
+		message: z.string(),
+		type: z.string().nullish(),
+		param: z.unknown().nullish(),
+		code: z.unknown().nullish()
+	}) }),
+	errorToMessage: (data) => data.error.message
+};
+/**
+* Recursively remove the JSON Schema `$schema` marker, which some gateway
+* backends (notably Gemini) reject in tool/structured-output schemas. Other
+* backends ignore its absence, so stripping it everywhere is safe.
+*/
+function stripJsonSchemaMarker(value) {
+	if (Array.isArray(value)) return value.map(stripJsonSchemaMarker);
+	if (value !== null && typeof value === "object") {
+		const result = {};
+		for (const [key, entry] of Object.entries(value)) {
+			if (key === "$schema") continue;
+			result[key] = stripJsonSchemaMarker(entry);
+		}
+		return result;
+	}
+	return value;
+}
+function transformNeonRequestBody(args) {
+	const transformed = { ...args };
+	if (transformed.tools != null) transformed.tools = stripJsonSchemaMarker(transformed.tools);
+	if (transformed.response_format != null) transformed.response_format = stripJsonSchemaMarker(transformed.response_format);
+	return transformed;
+}
+function createNeon(options = {}) {
+	const getHost = () => withoutTrailingSlash(loadSetting({
+		settingValue: options.baseURL,
+		environmentVariableName: "NEON_AI_GATEWAY_BASE_URL",
+		settingName: "baseURL",
+		description: "Neon AI Gateway base URL"
+	}));
+	const getHeaders = (extra) => withUserAgentSuffix({
+		Authorization: `Bearer ${loadApiKey({
+			apiKey: options.apiKey,
+			environmentVariableName: "NEON_AI_GATEWAY_TOKEN",
+			description: "Neon AI Gateway token"
+		})}`,
+		...extra,
+		...options.headers
+	}, `neondatabase/ai-sdk-provider/${VERSION}`);
+	const createAnthropicModel = (modelId) => new NeonAnthropicLanguageModel(modelId, {
+		provider: "neon.anthropic",
+		baseURL: `${getHost()}/ai-gateway/anthropic/v1`,
+		headers: () => getHeaders({ "anthropic-version": "2023-06-01" }),
+		fetch: options.fetch,
+		generateId
+	});
+	const createOpenAIModel = (modelId) => new NeonResponsesLanguageModel(modelId, {
+		provider: "neon.openai.responses",
+		url: ({ path }) => `${getHost()}/ai-gateway/openai/v1${path}`,
+		headers: getHeaders,
+		fetch: options.fetch,
+		fileIdPrefixes: ["file-"]
+	});
+	const createChatModel = (modelId) => new NeonChatLanguageModel(modelId, {
+		provider: "neon.chat",
+		url: ({ path }) => `${getHost()}/ai-gateway/mlflow/v1${path}`,
+		headers: getHeaders,
+		fetch: options.fetch,
+		errorStructure: neonErrorStructure,
+		transformRequestBody: transformNeonRequestBody,
+		supportsStructuredOutputs: true
+	});
+	const createLanguageModel = (modelId) => {
+		switch (getNeonModelRoute(modelId)) {
+			case "anthropic": return createAnthropicModel(modelId);
+			case "openai": return createOpenAIModel(modelId);
+			default: return createChatModel(modelId);
+		}
+	};
+	const provider = (modelId) => createLanguageModel(modelId);
+	provider.specificationVersion = "v3";
+	provider.languageModel = createLanguageModel;
+	provider.chat = createLanguageModel;
+	provider.tools = openai.tools;
+	provider.embeddingModel = (modelId) => {
+		throw new NoSuchModelError({
+			modelId,
+			modelType: "embeddingModel"
+		});
+	};
+	provider.textEmbeddingModel = provider.embeddingModel;
+	provider.imageModel = (modelId) => {
+		throw new NoSuchModelError({
+			modelId,
+			modelType: "imageModel"
+		});
+	};
+	return provider;
+}
+/** Default Neon provider instance (reads `NEON_AI_GATEWAY_*` env vars). */
+const neon = createNeon();
+//#endregion
+export { createNeon, neon };
+
+//# sourceMappingURL=provider.js.map

package/dist/lib/provider.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"provider.js","names":["openaiProvider"],"sources":["../../src/lib/provider.ts"],"sourcesContent":["/**\n * Community Vercel AI SDK provider for the Neon AI Gateway.\n *\n * `createNeon()` returns a provider that routes each model to the best gateway\n * endpoint based on its id (Anthropic → native Messages, OpenAI → native\n * Responses incl. Codex, everything else → unified MLflow), so a single\n * `neon('claude-...')` call (same base URL + token) reaches the whole catalog.\n * Ids use the canonical Neon (unprefixed) form; the legacy `databricks-` prefix\n * is also accepted. Configure with the branch-scoped `NEON_AI_GATEWAY_BASE_URL` +\n * `NEON_AI_GATEWAY_TOKEN` emitted by `neonctl env pull` / `neon dev`, or pass\n * `baseURL` / `apiKey` explicitly.\n */\nimport { openai as openaiProvider } from \"@ai-sdk/openai\";\nimport type { ProviderErrorStructure } from \"@ai-sdk/openai-compatible\";\nimport {\n\ttype LanguageModelV3,\n\tNoSuchModelError,\n\ttype ProviderV3,\n} from \"@ai-sdk/provider\";\nimport {\n\ttype FetchFunction,\n\tgenerateId,\n\tloadApiKey,\n\tloadSetting,\n\twithoutTrailingSlash,\n\twithUserAgentSuffix,\n} from \"@ai-sdk/provider-utils\";\nimport { z } from \"zod/v4\";\nimport { NeonAnthropicLanguageModel } from \"./neon-anthropic-language-model.js\";\nimport { NeonChatLanguageModel } from \"./neon-chat-language-model.js\";\nimport type { NeonChatModelId } from \"./neon-chat-options.js\";\nimport { getNeonModelRoute } from \"./neon-model-capabilities.js\";\nimport { NeonResponsesLanguageModel } from \"./neon-responses-language-model.js\";\nimport { VERSION } from \"./version.js\";\n\nconst neonErrorSchema = z.object({\n\terror: z.object({\n\t\tmessage: z.string(),\n\t\ttype: z.string().nullish(),\n\t\tparam: z.unknown().nullish(),\n\t\tcode: z.unknown().nullish(),\n\t}),\n});\n\nexport type NeonErrorData = z.infer<typeof neonErrorSchema>;\n\nconst neonErrorStructure: ProviderErrorStructure<NeonErrorData> = {\n\terrorSchema: neonErrorSchema,\n\terrorToMessage: (data) => data.error.message,\n};\n\n/**\n * Recursively remove the JSON Schema `$schema` marker, which some gateway\n * backends (notably Gemini) reject in tool/structured-output schemas. Other\n * backends ignore its absence, so stripping it everywhere is safe.\n */\nfunction stripJsonSchemaMarker(value: unknown): unknown {\n\tif (Array.isArray(value)) {\n\t\treturn value.map(stripJsonSchemaMarker);\n\t}\n\tif (value !== null && typeof value === \"object\") {\n\t\tconst result: Record<string, unknown> = {};\n\t\tfor (const [key, entry] of Object.entries(value)) {\n\t\t\tif (key === \"$schema\") {\n\t\t\t\tcontinue;\n\t\t\t}\n\t\t\tresult[key] = stripJsonSchemaMarker(entry);\n\t\t}\n\t\treturn result;\n\t}\n\treturn value;\n}\n\nfunction transformNeonRequestBody(\n\targs: Record<string, unknown>,\n): Record<string, unknown> {\n\tconst transformed = { ...args };\n\tif (transformed.tools != null) {\n\t\ttransformed.tools = stripJsonSchemaMarker(transformed.tools);\n\t}\n\tif (transformed.response_format != null) {\n\t\ttransformed.response_format = stripJsonSchemaMarker(\n\t\t\ttransformed.response_format,\n\t\t);\n\t}\n\treturn transformed;\n}\n\nexport interface NeonProviderSettings {\n\t/**\n\t * Neon AI Gateway base URL — the branch-scoped host root, e.g.\n\t * `https://<branch-id>-api.ai.<region>.aws.neon.tech`. The gateway paths are\n\t * appended internally. Falls back to the `NEON_AI_GATEWAY_BASE_URL` env var.\n\t */\n\tbaseURL?: string;\n\n\t/**\n\t * Neon AI Gateway platform token (the `nt_live_...` value). Falls back to the\n\t * `NEON_AI_GATEWAY_TOKEN` env var.\n\t */\n\tapiKey?: string;\n\n\t/** Custom headers to include in the requests. */\n\theaders?: Record<string, string>;\n\n\t/** Custom fetch implementation (e.g. for testing or middleware). */\n\tfetch?: FetchFunction;\n}\n\nexport interface NeonProvider extends ProviderV3 {\n\t/** Creates a Neon AI Gateway model for text generation. */\n\t(modelId: NeonChatModelId): LanguageModelV3;\n\n\t/** Creates a Neon AI Gateway model for text generation. */\n\tlanguageModel(modelId: NeonChatModelId): LanguageModelV3;\n\n\t/** Creates a Neon AI Gateway chat model for text generation. */\n\tchat(modelId: NeonChatModelId): LanguageModelV3;\n\n\t/** OpenAI Responses tools (e.g. `imageGeneration`) for OpenAI-routed models. */\n\ttools: typeof openaiProvider.tools;\n\n\t/** @deprecated Use `embeddingModel` instead. */\n\ttextEmbeddingModel(modelId: string): never;\n}\n\nexport function createNeon(options: NeonProviderSettings = {}): NeonProvider {\n\t// Resolved lazily so that `createNeon()` and the default `neon` instance do\n\t// not throw at import time when configuration comes from the environment.\n\tconst getHost = () =>\n\t\twithoutTrailingSlash(\n\t\t\tloadSetting({\n\t\t\t\tsettingValue: options.baseURL,\n\t\t\t\tenvironmentVariableName: \"NEON_AI_GATEWAY_BASE_URL\",\n\t\t\t\tsettingName: \"baseURL\",\n\t\t\t\tdescription: \"Neon AI Gateway base URL\",\n\t\t\t}),\n\t\t);\n\n\tconst getHeaders = (extra?: Record<string, string>) =>\n\t\twithUserAgentSuffix(\n\t\t\t{\n\t\t\t\tAuthorization: `Bearer ${loadApiKey({\n\t\t\t\t\tapiKey: options.apiKey,\n\t\t\t\t\tenvironmentVariableName: \"NEON_AI_GATEWAY_TOKEN\",\n\t\t\t\t\tdescription: \"Neon AI Gateway token\",\n\t\t\t\t})}`,\n\t\t\t\t...extra,\n\t\t\t\t...options.headers,\n\t\t\t},\n\t\t\t`neondatabase/ai-sdk-provider/${VERSION}`,\n\t\t);\n\n\t// Anthropic models -> native Messages API.\n\tconst createAnthropicModel = (modelId: NeonChatModelId) =>\n\t\tnew NeonAnthropicLanguageModel(modelId, {\n\t\t\tprovider: \"neon.anthropic\",\n\t\t\tbaseURL: `${getHost()}/ai-gateway/anthropic/v1`,\n\t\t\theaders: () => getHeaders({ \"anthropic-version\": \"2023-06-01\" }),\n\t\t\tfetch: options.fetch,\n\t\t\tgenerateId,\n\t\t});\n\n\t// OpenAI models (incl. Codex, only served natively) -> Responses API.\n\tconst createOpenAIModel = (modelId: NeonChatModelId) =>\n\t\tnew NeonResponsesLanguageModel(modelId, {\n\t\t\tprovider: \"neon.openai.responses\",\n\t\t\turl: ({ path }) => `${getHost()}/ai-gateway/openai/v1${path}`,\n\t\t\theaders: getHeaders,\n\t\t\tfetch: options.fetch,\n\t\t\tfileIdPrefixes: [\"file-\"],\n\t\t});\n\n\t// Everything else (Gemini, Llama, Qwen, gpt-oss, ...) -> unified MLflow\n\t// endpoint. Gemini is here because its native endpoint can't stream.\n\tconst createChatModel = (modelId: NeonChatModelId) =>\n\t\tnew NeonChatLanguageModel(modelId, {\n\t\t\tprovider: \"neon.chat\",\n\t\t\turl: ({ path }) => `${getHost()}/ai-gateway/mlflow/v1${path}`,\n\t\t\theaders: getHeaders,\n\t\t\tfetch: options.fetch,\n\t\t\terrorStructure: neonErrorStructure,\n\t\t\ttransformRequestBody: transformNeonRequestBody,\n\t\t\tsupportsStructuredOutputs: true,\n\t\t});\n\n\tconst createLanguageModel = (modelId: NeonChatModelId): LanguageModelV3 => {\n\t\tswitch (getNeonModelRoute(modelId)) {\n\t\t\tcase \"anthropic\":\n\t\t\t\treturn createAnthropicModel(modelId);\n\t\t\tcase \"openai\":\n\t\t\t\treturn createOpenAIModel(modelId);\n\t\t\tdefault:\n\t\t\t\treturn createChatModel(modelId);\n\t\t}\n\t};\n\n\tconst provider = (modelId: NeonChatModelId) => createLanguageModel(modelId);\n\n\tprovider.specificationVersion = \"v3\" as const;\n\tprovider.languageModel = createLanguageModel;\n\tprovider.chat = createLanguageModel;\n\tprovider.tools = openaiProvider.tools;\n\n\tprovider.embeddingModel = (modelId: string) => {\n\t\tthrow new NoSuchModelError({ modelId, modelType: \"embeddingModel\" });\n\t};\n\tprovider.textEmbeddingModel = provider.embeddingModel;\n\tprovider.imageModel = (modelId: string) => {\n\t\tthrow new NoSuchModelError({ modelId, modelType: \"imageModel\" });\n\t};\n\n\treturn provider;\n}\n\n/** Default Neon provider instance (reads `NEON_AI_GATEWAY_*` env vars). */\nexport const neon = createNeon();\n"],"mappings":";;;;;;;;;;;;;;;;;;;;;;AA8CA,MAAM,qBAA4D;CACjE,aAZuB,EAAE,OAAO,EAChC,OAAO,EAAE,OAAO;EACf,SAAS,EAAE,OAAO;EAClB,MAAM,EAAE,OAAO,CAAC,CAAC,QAAQ;EACzB,OAAO,EAAE,QAAQ,CAAC,CAAC,QAAQ;EAC3B,MAAM,EAAE,QAAQ,CAAC,CAAC,QAAQ;CAC3B,CAAC,EACF,CAK4B;CAC3B,iBAAiB,SAAS,KAAK,MAAM;AACtC;;;;;;AAOA,SAAS,sBAAsB,OAAyB;CACvD,IAAI,MAAM,QAAQ,KAAK,GACtB,OAAO,MAAM,IAAI,qBAAqB;CAEvC,IAAI,UAAU,QAAQ,OAAO,UAAU,UAAU;EAChD,MAAM,SAAkC,CAAC;EACzC,KAAK,MAAM,CAAC,KAAK,UAAU,OAAO,QAAQ,KAAK,GAAG;GACjD,IAAI,QAAQ,WACX;GAED,OAAO,OAAO,sBAAsB,KAAK;EAC1C;EACA,OAAO;CACR;CACA,OAAO;AACR;AAEA,SAAS,yBACR,MAC0B;CAC1B,MAAM,cAAc,EAAE,GAAG,KAAK;CAC9B,IAAI,YAAY,SAAS,MACxB,YAAY,QAAQ,sBAAsB,YAAY,KAAK;CAE5D,IAAI,YAAY,mBAAmB,MAClC,YAAY,kBAAkB,sBAC7B,YAAY,eACb;CAED,OAAO;AACR;AAwCA,SAAgB,WAAW,UAAgC,CAAC,GAAiB;CAG5E,MAAM,gBACL,qBACC,YAAY;EACX,cAAc,QAAQ;EACtB,yBAAyB;EACzB,aAAa;EACb,aAAa;CACd,CAAC,CACF;CAED,MAAM,cAAc,UACnB,oBACC;EACC,eAAe,UAAU,WAAW;GACnC,QAAQ,QAAQ;GAChB,yBAAyB;GACzB,aAAa;EACd,CAAC;EACD,GAAG;EACH,GAAG,QAAQ;CACZ,GACA,gCAAgC,SACjC;CAGD,MAAM,wBAAwB,YAC7B,IAAI,2BAA2B,SAAS;EACvC,UAAU;EACV,SAAS,GAAG,QAAQ,EAAE;EACtB,eAAe,WAAW,EAAE,qBAAqB,aAAa,CAAC;EAC/D,OAAO,QAAQ;EACf;CACD,CAAC;CAGF,MAAM,qBAAqB,YAC1B,IAAI,2BAA2B,SAAS;EACvC,UAAU;EACV,MAAM,EAAE,WAAW,GAAG,QAAQ,EAAE,uBAAuB;EACvD,SAAS;EACT,OAAO,QAAQ;EACf,gBAAgB,CAAC,OAAO;CACzB,CAAC;CAIF,MAAM,mBAAmB,YACxB,IAAI,sBAAsB,SAAS;EAClC,UAAU;EACV,MAAM,EAAE,WAAW,GAAG,QAAQ,EAAE,uBAAuB;EACvD,SAAS;EACT,OAAO,QAAQ;EACf,gBAAgB;EAChB,sBAAsB;EACtB,2BAA2B;CAC5B,CAAC;CAEF,MAAM,uBAAuB,YAA8C;EAC1E,QAAQ,kBAAkB,OAAO,GAAjC;GACC,KAAK,aACJ,OAAO,qBAAqB,OAAO;GACpC,KAAK,UACJ,OAAO,kBAAkB,OAAO;GACjC,SACC,OAAO,gBAAgB,OAAO;EAChC;CACD;CAEA,MAAM,YAAY,YAA6B,oBAAoB,OAAO;CAE1E,SAAS,uBAAuB;CAChC,SAAS,gBAAgB;CACzB,SAAS,OAAO;CAChB,SAAS,QAAQA,OAAe;CAEhC,SAAS,kBAAkB,YAAoB;EAC9C,MAAM,IAAI,iBAAiB;GAAE;GAAS,WAAW;EAAiB,CAAC;CACpE;CACA,SAAS,qBAAqB,SAAS;CACvC,SAAS,cAAc,YAAoB;EAC1C,MAAM,IAAI,iBAAiB;GAAE;GAAS,WAAW;EAAa,CAAC;CAChE;CAEA,OAAO;AACR;;AAGA,MAAa,OAAO,WAAW"}

package/dist/lib/version.d.ts

@@ -0,0 +1,5 @@
+//#region src/lib/version.d.ts
+declare const VERSION: string;
+//#endregion
+export { VERSION };
+//# sourceMappingURL=version.d.ts.map

package/dist/lib/version.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"version.d.ts","names":[],"sources":["../../src/lib/version.ts"],"mappings":";cAIa"}

package/dist/lib/version.js

@@ -0,0 +1,7 @@
+import { version } from "../package.js";
+//#region src/lib/version.ts
+const VERSION = version;
+//#endregion
+export { VERSION };
+
+//# sourceMappingURL=version.js.map

package/dist/lib/version.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"version.js","names":["pkg.version"],"sources":["../../src/lib/version.ts"],"sourcesContent":["import pkg from \"../../package.json\" with { type: \"json\" };\n\n// Package version, used in the User-Agent suffix sent to the gateway. Derived\n// from package.json so it stays in sync with the changeset version bump flow.\nexport const VERSION: string = pkg.version;\n"],"mappings":";;AAIA,MAAa,UAAkBA"}

package/dist/package.js

@@ -0,0 +1,6 @@
+//#region package.json
+var version = "0.6.0";
+//#endregion
+export { version };
+
+//# sourceMappingURL=package.js.map

package/dist/package.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"package.js","names":[],"sources":["../package.json"],"sourcesContent":[""],"mappings":""}

package/package.json

@@ -1,30 +1,74 @@
 {
   "name": "@neon/ai-sdk-provider",
-  "version": "0.0.0",
-  "description": "Neon provider for the AI SDK.",
+  "version": "0.6.0",
+  "description": "Community Vercel AI SDK provider for the Neon AI Gateway.",
+  "keywords": [
+    "neon",
+    "ai",
+    "ai-sdk",
+    "vercel-ai-sdk",
+    "provider",
+    "ai-gateway",
+    "llm",
+    "platform"
+  ],
+  "repository": {
+    "type": "git",
+    "url": "git+https://github.com/neondatabase/neon-pkgs.git"
+  },
   "license": "Apache-2.0",
   "author": {
     "name": "Neon",
     "url": "https://neon.com"
   },
-  "repository": {
-    "type": "git",
-    "url": "git+https://github.com/neondatabase/neon-pkgs.git",
-    "directory": "packages/ai-sdk-provider"
-  },
-  "homepage": "https://github.com/neondatabase/neon-pkgs",
-  "bugs": {
-    "url": "https://github.com/neondatabase/neon-pkgs/issues"
+  "type": "module",
+  "main": "dist/index.js",
+  "types": "dist/index.d.ts",
+  "exports": {
+    ".": {
+      "types": "./dist/index.d.ts",
+      "import": "./dist/index.js",
+      "default": "./dist/index.js"
+    }
   },
-  "keywords": [
-    "neon",
-    "ai",
-    "ai-sdk",
-    "database",
-    "postgres"
+  "files": [
+    "README.md",
+    "dist/",
+    "package.json"
   ],
+  "devDependencies": {
+    "@types/node": "^20.19.0",
+    "@vitest/coverage-v8": "3.0.9",
+    "ai": "6.0.132",
+    "console-fail-test": "0.5.0",
+    "tsdown": "^0.14.1",
+    "typescript": "^5.8.2",
+    "vitest": "^3.0.9",
+    "zod": "^4.4.3"
+  },
+  "engines": {
+    "node": ">=20.19.0"
+  },
   "publishConfig": {
-    "access": "public",
     "provenance": false
+  },
+  "dependencies": {
+    "@ai-sdk/anthropic": "^3.0.82",
+    "@ai-sdk/openai": "^3.0.69",
+    "@ai-sdk/openai-compatible": "^2.0.48",
+    "@ai-sdk/provider": "^3.0.10",
+    "@ai-sdk/provider-utils": "^4.0.27"
+  },
+  "peerDependencies": {
+    "ai": "^6.0.0",
+    "zod": "^3.25.76 || ^4.1.8"
+  },
+  "scripts": {
+    "build": "tsc --noEmit && tsdown",
+    "test": "vitest --passWithNoTests",
+    "test:ci": "vitest run --passWithNoTests",
+    "test:drift": "NEON_DRIFT_CHECK=1 vitest run neon-catalog-drift",
+    "test:e2e": "vitest run --config vitest.e2e.config.ts",
+    "tsc": "tsc"
   }
-}
+}