@juspay/neurolink 12.49.4 → 12.49.6

package/dist/constants/contextWindows.d.ts

@@ -56,6 +56,17 @@ export declare function getRuntimeOutputCeiling(provider: string, model?: string
 export declare function clearRuntimeOutputCeiling(provider: string, model: string): void;
 /** Test hook: clear runtime-discovered output ceilings (state is module-global). */
 export declare function clearRuntimeOutputCeilings(): void;
+/**
+ * `us.anthropic.claude-sonnet-4-6` -> `anthropic.claude-sonnet-4-6`.
+ *
+ * Undefined for any other shape, so an id that merely starts with `us.` is
+ * never rewritten. Removes one geography at most: no vendor is a geography.
+ *
+ * Every Bedrock table keyed by bare ids (context windows here, output ceilings
+ * in `getSafeMaxTokens`) goes through this one function, so they cannot
+ * disagree about which ids are cross-region profile ids.
+ */
+export declare function stripBedrockGeoPrefix(model: string): string | undefined;
 /**
  * Resolve context window size for a provider/model combination.
  *

package/dist/constants/contextWindows.js

@@ -507,8 +507,12 @@ const BEDROCK_VENDORS = new Set(Object.values(BedrockModels)
  *
  * Undefined for any other shape, so an id that merely starts with `us.` is
  * never rewritten. Removes one geography at most: no vendor is a geography.
+ *
+ * Every Bedrock table keyed by bare ids (context windows here, output ceilings
+ * in `getSafeMaxTokens`) goes through this one function, so they cannot
+ * disagree about which ids are cross-region profile ids.
  */
-function stripBedrockGeoPrefix(model) {
+export function stripBedrockGeoPrefix(model) {
     const [geo, vendor, ...rest] = model.split(".");
     if (rest.length === 0 ||
         !BEDROCK_GEO_PREFIXES.has(geo) ||

package/dist/providers/litellm/client.d.ts

@@ -92,5 +92,10 @@ export declare class LiteLLMProvider extends OpenAIChatCompletionsProvider {
      * Generate embeddings for multiple text inputs via native /v1/embeddings.
      */
     embedMany(texts: string[], modelName?: string): Promise<number[][]>;
+    /**
+     * One non-empty vector per input, or fail: a dropped or empty row would
+     * pair later texts with the wrong vectors, or score 0 downstream.
+     */
+    private requireEmbeddings;
     private callEmbeddings;
 }

package/dist/providers/litellm/client.js

@@ -474,7 +474,8 @@ export class LiteLLMProvider extends OpenAIChatCompletionsProvider {
         const embeddingModelName = modelName ||
             process.env.LITELLM_EMBEDDING_MODEL ||
             "gemini-embedding-001";
-        const [embedding] = await this.callEmbeddings(embeddingModelName, [text], "embed");
+        const rows = await this.callEmbeddings(embeddingModelName, [text], "embed");
+        const [embedding] = this.requireEmbeddings(rows, 1, embeddingModelName);
         return embedding;
     }
     /**
@@ -484,7 +485,34 @@ export class LiteLLMProvider extends OpenAIChatCompletionsProvider {
         const embeddingModelName = modelName ||
             process.env.LITELLM_EMBEDDING_MODEL ||
             "gemini-embedding-001";
-        return this.callEmbeddings(embeddingModelName, texts, "embedMany");
+        const rows = await this.callEmbeddings(embeddingModelName, texts, "embedMany");
+        return this.requireEmbeddings(rows, texts.length, embeddingModelName);
+    }
+    /**
+     * One non-empty vector per input, or fail: a dropped or empty row would
+     * pair later texts with the wrong vectors, or score 0 downstream.
+     */
+    requireEmbeddings(rows, inputCount, modelName) {
+        if (rows.length !== inputCount) {
+            logger.error("Embedding response held the wrong number of vectors", {
+                provider: this.providerName,
+                model: modelName,
+                expected: inputCount,
+                received: rows.length,
+            });
+            throw new ProviderError(`Embedding response held ${rows.length} vectors for ${inputCount} texts`, this.providerName);
+        }
+        return rows.map((values, index) => {
+            if (!values || values.length === 0) {
+                logger.error("Embedding returned no values", {
+                    provider: this.providerName,
+                    model: modelName,
+                    index,
+                });
+                throw new ProviderError(`Embedding for the text at index ${index} came back without values`, this.providerName);
+            }
+            return values;
+        });
     }
     async callEmbeddings(modelName, input, operation) {
         const url = `${stripTrailingSlash(this.config.baseURL)}/embeddings`;
@@ -514,13 +542,11 @@ export class LiteLLMProvider extends OpenAIChatCompletionsProvider {
                     `LiteLLM ${operation} failed with status ${res.status}`));
             }
             const json = (await res.json());
-            const embeddings = (json.data ?? [])
-                .map((row) => row.embedding)
-                .filter((e) => Array.isArray(e));
-            if (embeddings.length === 0) {
+            const rows = (json.data ?? []).map((row) => Array.isArray(row.embedding) ? row.embedding : undefined);
+            if (rows.length === 0) {
                 throw new ProviderError(`LiteLLM ${operation} returned no embeddings`, this.providerName);
             }
-            return embeddings;
+            return rows;
         }
         finally {
             timeoutController?.cleanup();

package/dist/providers/ollama/client.d.ts

@@ -67,5 +67,10 @@ export declare class OllamaProvider extends OpenAIChatCompletionsProvider {
      * Generate embeddings for multiple text inputs via native /v1/embeddings.
      */
     embedMany(texts: string[], modelName?: string): Promise<number[][]>;
+    /**
+     * One non-empty vector per input, or fail: a dropped or empty row would
+     * pair later texts with the wrong vectors, or score 0 downstream.
+     */
+    private requireEmbeddings;
     private callEmbeddings;
 }

package/dist/providers/ollama/client.js

@@ -216,7 +216,8 @@ export class OllamaProvider extends OpenAIChatCompletionsProvider {
         const embeddingModel = modelName ||
             process.env.OLLAMA_EMBEDDING_MODEL ||
             DEFAULT_EMBEDDING_MODEL;
-        const [embedding] = await this.callEmbeddings(embeddingModel, [text], "embed");
+        const rows = await this.callEmbeddings(embeddingModel, [text], "embed");
+        const [embedding] = this.requireEmbeddings(rows, 1, embeddingModel);
         return embedding;
     }
     /**
@@ -226,7 +227,34 @@ export class OllamaProvider extends OpenAIChatCompletionsProvider {
         const embeddingModel = modelName ||
             process.env.OLLAMA_EMBEDDING_MODEL ||
             DEFAULT_EMBEDDING_MODEL;
-        return this.callEmbeddings(embeddingModel, texts, "embedMany");
+        const rows = await this.callEmbeddings(embeddingModel, texts, "embedMany");
+        return this.requireEmbeddings(rows, texts.length, embeddingModel);
+    }
+    /**
+     * One non-empty vector per input, or fail: a dropped or empty row would
+     * pair later texts with the wrong vectors, or score 0 downstream.
+     */
+    requireEmbeddings(rows, inputCount, modelName) {
+        if (rows.length !== inputCount) {
+            logger.error("Embedding response held the wrong number of vectors", {
+                provider: this.providerName,
+                model: modelName,
+                expected: inputCount,
+                received: rows.length,
+            });
+            throw new ProviderError(`Embedding response held ${rows.length} vectors for ${inputCount} texts`, this.providerName);
+        }
+        return rows.map((values, index) => {
+            if (!values || values.length === 0) {
+                logger.error("Embedding returned no values", {
+                    provider: this.providerName,
+                    model: modelName,
+                    index,
+                });
+                throw new ProviderError(`Embedding for the text at index ${index} came back without values`, this.providerName);
+            }
+            return values;
+        });
     }
     async callEmbeddings(modelName, input, operation) {
         const url = `${stripTrailingSlash(this.config.baseURL)}/embeddings`;
@@ -265,13 +293,11 @@ export class OllamaProvider extends OpenAIChatCompletionsProvider {
                     `Ollama ${operation} failed with status ${res.status}`));
             }
             const json = (await res.json());
-            const embeddings = (json.data ?? [])
-                .map((row) => row.embedding)
-                .filter((e) => Array.isArray(e));
-            if (embeddings.length === 0) {
+            const rows = (json.data ?? []).map((row) => Array.isArray(row.embedding) ? row.embedding : undefined);
+            if (rows.length === 0) {
                 throw new ProviderError(`Ollama ${operation} returned no embeddings`, "ollama");
             }
-            return embeddings;
+            return rows;
         }
         finally {
             timeoutController?.cleanup();

package/dist/proxy/utils/noProxyUtils.d.ts

@@ -17,6 +17,8 @@
  * - "192.168.1.10", "::1", "[::1]:8080" - an IP literal, exactly
  * - "192.168.1.0/24" - an IPv4 CIDR range, for IPv4 literal targets (IPv6 CIDR
  *   ranges are not supported)
+ * - "." and ".." - an entry made only of dots names no host, so it matches
+ *   nothing, as in curl
  *
  * @param targetUrl - The URL to check for proxy bypass
  * @param noProxyEnv - Optional NO_PROXY environment variable value (if not provided, reads from process.env)
@@ -37,6 +39,8 @@ export declare function getNoProxyEnv(): string | undefined;
  * - "example.com" - exact match
  * - ".example.com" - domain suffix match
  *
+ * An entry made only of dots names no host and matches nothing.
+ *
  * @param targetUrl - The URL to check
  * @param noProxyValue - The NO_PROXY value to check against
  * @returns true if should bypass proxy

package/dist/proxy/utils/noProxyUtils.js

@@ -44,6 +44,10 @@ function canonicalIpv6Host(address) {
 function withoutTrailingDot(name) {
     return name.endsWith(".") ? name.slice(0, -1) : name;
 }
+/** An entry made only of dots (`.`, `..`) names no host. */
+function isDotsOnly(entry) {
+    return /^\.+$/.test(entry);
+}
 /**
  * curl's rules for one list entry: an optional leading `.` or `*.` is ignored;
  * the entry names that host and every subdomain of it, an IP literal only
@@ -106,13 +110,20 @@ function matchesNoProxyEntry(entry, hostname, port) {
     return name === domain || name.endsWith(`.${domain}`);
 }
 /**
- * The rules this function applied before it followed curl, unchanged and still
- * consulted, so that a configuration that bypassed the proxy through them keeps
- * doing so. That includes where they are looser than curl (`.example.com` also
- * matches `notexample.com`) and an IPv4 CIDR range such as `192.168.1.0/24`,
- * which curl-style entries do not express.
+ * The rules this function applied before it followed curl, still consulted so
+ * that a configuration that bypassed the proxy through them keeps doing so.
+ * That includes where they are looser than curl (`.example.com` also matches
+ * `notexample.com`) and an IPv4 CIDR range such as `192.168.1.0/24`, which
+ * curl-style entries do not express. They are unchanged except that an entry
+ * made only of dots names no host, as in the curl-style rules.
  */
 function matchesLegacyNoProxyEntry(lowerPattern, hostname, port) {
+    // Taken as a suffix, "." leaves the empty string, which every host name ends
+    // with, and ".." leaves ".", which every host written with a trailing dot
+    // ends with.
+    if (isDotsOnly(lowerPattern)) {
+        return false;
+    }
     // Domain suffix match (.example.com)
     if (lowerPattern.startsWith(".")) {
         const suffix = lowerPattern.slice(1);
@@ -145,6 +156,8 @@ function matchesLegacyNoProxyEntry(lowerPattern, hostname, port) {
  * - "192.168.1.10", "::1", "[::1]:8080" - an IP literal, exactly
  * - "192.168.1.0/24" - an IPv4 CIDR range, for IPv4 literal targets (IPv6 CIDR
  *   ranges are not supported)
+ * - "." and ".." - an entry made only of dots names no host, so it matches
+ *   nothing, as in curl
  *
  * @param targetUrl - The URL to check for proxy bypass
  * @param noProxyEnv - Optional NO_PROXY environment variable value (if not provided, reads from process.env)
@@ -197,6 +210,8 @@ export function getNoProxyEnv() {
  * - "example.com" - exact match
  * - ".example.com" - domain suffix match
  *
+ * An entry made only of dots names no host and matches nothing.
+ *
  * @param targetUrl - The URL to check
  * @param noProxyValue - The NO_PROXY value to check against
  * @returns true if should bypass proxy
@@ -211,6 +226,10 @@ export function shouldBypassProxySimple(targetUrl, noProxyValue) {
             if (!pattern) {
                 continue;
             }
+            // As a suffix, "." would match every host written with a trailing dot.
+            if (isDotsOnly(pattern)) {
+                continue;
+            }
             // Exact match
             if (hostname === pattern) {
                 return true;

package/dist/types/providers.d.ts

@@ -136,6 +136,13 @@ export type NeurolinkCredentials = {
         /** Inline service-account fields (alternative to serviceAccountKey) */
         clientEmail?: string;
         privateKey?: string;
+        /**
+         * Overrides the Vertex API base URL, e.g. to reach a local stand-in. Used
+         * for Gemini and Claude models, in Express Mode and with a project and
+         * location. Takes precedence over `GOOGLE_VERTEX_BASE_URL`; a blank value
+         * counts as unset.
+         */
+        baseURL?: string;
     };
     bedrock?: {
         accessKeyId?: string;

package/dist/utils/tokenLimits.js

@@ -2,7 +2,7 @@
  * Provider-specific token limit utilities
  * Provides safe maxTokens values based on provider and model capabilities
  */
-import { getRuntimeOutputCeiling } from "../constants/contextWindows.js";
+import { getRuntimeOutputCeiling, stripBedrockGeoPrefix, } from "../constants/contextWindows.js";
 import { PROVIDER_MAX_TOKENS } from "../core/constants.js";
 import { logger } from "./logger.js";
 import { hasRestrictedOutputLimit, RESTRICTED_OUTPUT_TOKEN_LIMIT, } from "./modelDetection.js";
@@ -67,6 +67,11 @@ export function getSafeMaxTokens(provider, model, requestedMaxTokens) {
         }
         return requestedMaxTokens;
     }
+    // The Bedrock rows are keyed by bare id, but for many models the cross-region
+    // profile id (us.amazon.nova-pro-v1:0) is the only one a caller can invoke.
+    // The bare id is tried only after the id as given misses, so an explicit row
+    // for the prefixed id still wins.
+    const bareModel = provider === "bedrock" && model ? stripBedrockGeoPrefix(model) : undefined;
     // Get model-specific limit or provider default
     let maxLimit;
     if (model &&
@@ -74,6 +79,11 @@ export function getSafeMaxTokens(provider, model, requestedMaxTokens) {
         providerLimits[model]) {
         maxLimit = providerLimits[model];
     }
+    else if (bareModel !== undefined &&
+        typeof providerLimits === "object" &&
+        providerLimits[bareModel]) {
+        maxLimit = providerLimits[bareModel];
+    }
     else if (typeof providerLimits === "object" &&
         providerLimits.default) {
         maxLimit = providerLimits.default;

package/docs-site/static/search-index.json

@@ -4146,13 +4146,13 @@
 {"objectID":"1d8625a67cba7bc640e454fe15f76d86e2be4fbe79e9496fc085e5a6df496086","title":"Future Enhancements","url":"/docs/features/pdf-support#future-enhancements","content":"Planned features for PDF support:\nOCR Integration: Extract text from scanned PDFs\nPage Selection: Analyze specific pages only\nPDF Generation: Create PDFs from AI responses\nForm Filling: Extract and populate PDF forms","hierarchy":{"lvl0":"Features","lvl1":"PDF File Support","lvl2":"Future Enhancements","lvl3":""}},
 {"objectID":"20bb5119cb74199a04bebd17c621d8c6a0438c0833f12103b9f3914c5f026ceb","title":"Feedback and Support","url":"/docs/features/pdf-support#feedback-and-support","content":"Found a bug or have a feature request? Please:\nCheck existing issues on GitHub\nCreate a new issue with:\nProvider used\nPDF file details (size, pages)\nError message or unexpected behavior\nSample code (if possible)","hierarchy":{"lvl0":"Features","lvl1":"PDF File Support","lvl2":"Feedback and Support","lvl3":""}},
 {"objectID":"7c47c75b9023d438078ad3ed696c167cb9b5beaa369fad844635f7f32cce74a5","title":"Version 9.2.0 (Current)","url":"/docs/features/pdf-support#version-920-current","content":"✅ Initial PDF support for Vertex AI, Anthropic, Bedrock, AI Studio\n✅ Auto-detection via --file flag\n✅ Multiple PDF processing\n✅ Size and page limit validation\n✅ Comprehensive error messages\n✅ CLI and SDK integration\n✅ Streaming support\n✅ Mixed multimodal inputs (PDF + CSV + images)\n\nNext: Multimodal Chat Guide | CSV Support","hierarchy":{"lvl0":"Features","lvl1":"PDF File Support","lvl2":"Version 9.2.0 (Current)","lvl3":""}},
-{"objectID":"2a46dad649fcc5c7b389d9db519211899183703302b4127f0a4a2285018e1362","title":"Per-Request Credentials","url":"/docs/features/per-request-credentials","content":"Per-Request Credentials\n\nStatus: Stable | Availability: SDK only\n\nOverview\n\nNeuroLink allows provider credentials to be supplied at two levels below the environment-variable default: on the NeuroLink constructor (instance level) and on individual generate() / stream() calls (per-call level). Credentials are resolved in the following order of precedence:\n\nThis enables multi-tenant architectures where different users or tenants supply their own provider API keys (bring-your-own-key, BYOK), without requiring separate NeuroLink instances per user or touching the process environment.\n\nTypical use cases:\nMulti-tenant SaaS — each API request carries the calling user's provider key; no shared key leakage between tenants\nBYOK products — end users paste their OpenAI / Anthropic keys in settings; you forward them to NeuroLink per call\nTesting and CI — inject credentials programmatically without setting environment variables\nProvider switching — override only the active provider's credentials while leaving others to fall through to env vars\n\nQuick Start\n\nInstance-Level vs Per-Call\n\nInstance-Level Credentials\n\nSet credentials in the NeuroLink constructor. These apply as the default for every generate() and stream() call made on that instance. Useful when serving a single tenant or when you have a known key for the duration of the instance lifecycle.\n\nPer-Call Credentials\n\nSet credentials directly on generate() or stream(). These override the instance-level credentials for that single call only. Only the providers you explicitly set are overridden — others continue falling through to instance credentials and then environment variables.\n\nPrecedence Rules\n\n| Level       | Scope                 | Set on                                                                              |\n| ----------- | --------------------- | ----------------------------------------------------------------------------------- |\n| Per-call    | Single request only   | generate({ credentials }), stream({ credentials }) or decide({ credentials }) |\n| Instance    | All calls on instance | new NeuroLink({ credentials })                                                    |\n| Environment | Process-wide fallback | OPENAI_API_KEY, ANTHROPIC_API_KEY, …                                            |\n\nUnset providers at any level fall through to the next. You never need to repeat a credential at the per-call level if the instance default is correct.\n\nProvider Credential Reference\n\nAll fields are optional — omit any field you want to fall through to a lower-precedence level.\n\n| Provider             | Key                 | Fields                                                                                               |\n| -------------------- | ------------------- | ---------------------------------------------------------------------------------------------------- |\n| OpenAI               | openai            | apiKey, baseURL                                                                                  |\n| Anthropic            | anthropic         | apiKey, oauthToken, baseURL                                                                    |\n| Google AI Studio     | googleAiStudio    | apiKey, baseURL                                                                                  |\n| Google Vertex AI     | vertex            | projectId, location, apiKey (Express Mode), serviceAccountKey, clientEmail, privateKey   |\n| Amazon Bedrock       | bedrock           | accessKeyId, secretAccessKey, sessionToken, region                                           |\n| Amazon SageMaker     | sagemaker         | accessKeyId, secretAccessKey, sessionToken, region, endpoint                               |\n| Azure OpenAI         | azure             | apiKey, resourceName, deploymentName, apiVersion                                             |\n| Mistral              | mistral           | apiKey                                                                                             |\n| Hugging Face         | huggingFace       | apiKey, baseURL                                                                                  |\n| OpenRouter           | openrouter        | apiKey, baseURL                                                                                  |\n| LiteLLM              | litellm           | apiKey, baseURL                                                                                  |\n| OpenAI-Compatible    | openaiCompatible  | apiKey, baseURL                                                                                  |\n| Cerebras             | cerebras          | apiKey, baseURL                                                                                  |\n| SambaNova            | sambanova         | apiKey, baseURL                                                                                  |\n| Ollama               | ollama            | baseURL                                                                                       ","hierarchy":{"lvl0":"Features","lvl1":"Per-Request Credentials","lvl2":"","lvl3":""}},
+{"objectID":"2a46dad649fcc5c7b389d9db519211899183703302b4127f0a4a2285018e1362","title":"Per-Request Credentials","url":"/docs/features/per-request-credentials","content":"Per-Request Credentials\n\nStatus: Stable | Availability: SDK only\n\nOverview\n\nNeuroLink allows provider credentials to be supplied at two levels below the environment-variable default: on the NeuroLink constructor (instance level) and on individual generate() / stream() calls (per-call level). Credentials are resolved in the following order of precedence:\n\nThis enables multi-tenant architectures where different users or tenants supply their own provider API keys (bring-your-own-key, BYOK), without requiring separate NeuroLink instances per user or touching the process environment.\n\nTypical use cases:\nMulti-tenant SaaS — each API request carries the calling user's provider key; no shared key leakage between tenants\nBYOK products — end users paste their OpenAI / Anthropic keys in settings; you forward them to NeuroLink per call\nTesting and CI — inject credentials programmatically without setting environment variables\nProvider switching — override only the active provider's credentials while leaving others to fall through to env vars\n\nQuick Start\n\nInstance-Level vs Per-Call\n\nInstance-Level Credentials\n\nSet credentials in the NeuroLink constructor. These apply as the default for every generate() and stream() call made on that instance. Useful when serving a single tenant or when you have a known key for the duration of the instance lifecycle.\n\nPer-Call Credentials\n\nSet credentials directly on generate() or stream(). These override the instance-level credentials for that single call only. Only the providers you explicitly set are overridden — others continue falling through to instance credentials and then environment variables.\n\nPrecedence Rules\n\n| Level       | Scope                 | Set on                                                                              |\n| ----------- | --------------------- | ----------------------------------------------------------------------------------- |\n| Per-call    | Single request only   | generate({ credentials }), stream({ credentials }) or decide({ credentials }) |\n| Instance    | All calls on instance | new NeuroLink({ credentials })                                                    |\n| Environment | Process-wide fallback | OPENAI_API_KEY, ANTHROPIC_API_KEY, …                                            |\n\nUnset providers at any level fall through to the next. You never need to repeat a credential at the per-call level if the instance default is correct.\n\nProvider Credential Reference\n\nAll fields are optional — omit any field you want to fall through to a lower-precedence level.\n\n| Provider             | Key                 | Fields                                                                                                        |\n| -------------------- | ------------------- | ------------------------------------------------------------------------------------------------------------- |\n| OpenAI               | openai            | apiKey, baseURL                                                                                           |\n| Anthropic            | anthropic         | apiKey, oauthToken, baseURL                                                                             |\n| Google AI Studio     | googleAiStudio    | apiKey, baseURL                                                                                           |\n| Google Vertex AI     | vertex            | projectId, location, apiKey (Express Mode), serviceAccountKey, clientEmail, privateKey, baseURL |\n| Amazon Bedrock       | bedrock           | accessKeyId, secretAccessKey, sessionToken, region                                                    |\n| Amazon SageMaker     | sagemaker         | accessKeyId, secretAccessKey, sessionToken, region, endpoint                                        |\n| Azure OpenAI         | azure             | apiKey, resourceName, deploymentName, apiVersion                                                      |\n| Mistral              | mistral           | apiKey                                                                                                      |\n| Hugging Face         | huggingFace       | apiKey, baseURL                                                                                           |\n| OpenRouter           | openrouter        | apiKey, baseURL                                                                                           |\n| LiteLLM              | litellm           | apiKey, baseURL                                                                                           |\n| OpenAI-Compatible    | openaiCompatible  | apiKey, baseURL                                                                                           |\n| Cerebras             | cerebras          | apiKey, baseURL                                                                                           |\n| SambaNova            | sambanova         | apiKey, baseURL                                                                                          ","hierarchy":{"lvl0":"Features","lvl1":"Per-Request Credentials","lvl2":"","lvl3":""}},
 {"objectID":"12291afcfaaf4af65b01b07e93a9eac08280ff7c61394f6051a86c1b152a7b4a","title":"Per-Request Credentials","url":"/docs/features/per-request-credentials#per-request-credentials","content":"Status: Stable | Availability: SDK only","hierarchy":{"lvl0":"Features","lvl1":"Per-Request Credentials","lvl2":"Per-Request Credentials","lvl3":""}},
 {"objectID":"6c6756b0806344240b3c86d6b0063587db50f2c4ccf936a32e87d4377b77014c","title":"Overview","url":"/docs/features/per-request-credentials#overview","content":"NeuroLink allows provider credentials to be supplied at two levels below the environment-variable default: on the NeuroLink constructor (instance level) and on individual generate() / stream() calls (per-call level). Credentials are resolved in the following order of precedence:\n\nThis enables multi-tenant architectures where different users or tenants supply their own provider API keys (bring-your-own-key, BYOK), without requiring separate NeuroLink instances per user or touching the process environment.\n\nTypical use cases:\nMulti-tenant SaaS — each API request carries the calling user's provider key; no shared key leakage between tenants\nBYOK products — end users paste their OpenAI / Anthropic keys in settings; you forward them to NeuroLink per call\nTesting and CI — inject credentials programmatically without setting environment variables\nProvider switching — override only the active provider's credentials while leaving others to fall through to env vars","hierarchy":{"lvl0":"Features","lvl1":"Per-Request Credentials","lvl2":"Overview","lvl3":""}},
 {"objectID":"d4e86918319aa78b4e4461dbf29a60ebe90935e74fc70617f59bf524ba3e1eae","title":"Instance-Level Credentials","url":"/docs/features/per-request-credentials#instance-level-credentials","content":"Set credentials in the NeuroLink constructor. These apply as the default for every generate() and stream() call made on that instance. Useful when serving a single tenant or when you have a known key for the duration of the instance lifecycle.","hierarchy":{"lvl0":"Features","lvl1":"Per-Request Credentials","lvl2":"Instance-Level Credentials","lvl3":""}},
 {"objectID":"250b1dd11e018f5f252b27ce408e3dade21eff1cf26eed4aaf81126fad6ab691","title":"Per-Call Credentials","url":"/docs/features/per-request-credentials#per-call-credentials","content":"Set credentials directly on generate() or stream(). These override the instance-level credentials for that single call only. Only the providers you explicitly set are overridden — others continue falling through to instance credentials and then environment variables.","hierarchy":{"lvl0":"Features","lvl1":"Per-Request Credentials","lvl2":"Per-Call Credentials","lvl3":""}},
 {"objectID":"4166cbaf6e6263ac98e24d7ede9950b44aade18e787b57e71138d2fc2deaaa07","title":"Precedence Rules","url":"/docs/features/per-request-credentials#precedence-rules","content":"| Level       | Scope                 | Set on                                                                              |\n| ----------- | --------------------- | ----------------------------------------------------------------------------------- |\n| Per-call    | Single request only   | generate({ credentials }), stream({ credentials }) or decide({ credentials }) |\n| Instance    | All calls on instance | new NeuroLink({ credentials })                                                    |\n| Environment | Process-wide fallback | OPENAI_API_KEY, ANTHROPIC_API_KEY, …                                            |\n\nUnset providers at any level fall through to the next. You never need to repeat a credential at the per-call level if the instance default is correct.","hierarchy":{"lvl0":"Features","lvl1":"Per-Request Credentials","lvl2":"Precedence Rules","lvl3":""}},
-{"objectID":"b739fcf7819dd6f055ebe3bcdb161ff839067dc172235f9d0072f9cf79d7bcd3","title":"Provider Credential Reference","url":"/docs/features/per-request-credentials#provider-credential-reference","content":"All fields are optional — omit any field you want to fall through to a lower-precedence level.\n\n| Provider             | Key                 | Fields                                                                                               |\n| -------------------- | ------------------- | ---------------------------------------------------------------------------------------------------- |\n| OpenAI               | openai            | apiKey, baseURL                                                                                  |\n| Anthropic            | anthropic         | apiKey, oauthToken, baseURL                                                                    |\n| Google AI Studio     | googleAiStudio    | apiKey, baseURL                                                                                  |\n| Google Vertex AI     | vertex            | projectId, location, apiKey (Express Mode), serviceAccountKey, clientEmail, privateKey   |\n| Amazon Bedrock       | bedrock           | accessKeyId, secretAccessKey, sessionToken, region                                           |\n| Amazon SageMaker     | sagemaker         | accessKeyId, secretAccessKey, sessionToken, region, endpoint                               |\n| Azure OpenAI         | azure             | apiKey, resourceName, deploymentName, apiVersion                                             |\n| Mistral              | mistral           | apiKey                                                                                             |\n| Hugging Face         | huggingFace       | apiKey, baseURL                                                                                  |\n| OpenRouter           | openrouter        | apiKey, baseURL                                                                                  |\n| LiteLLM              | litellm           | apiKey, baseURL                                                                                  |\n| OpenAI-Compatible    | openaiCompatible","hierarchy":{"lvl0":"Features","lvl1":"Per-Request Credentials","lvl2":"Provider Credential Reference","lvl3":""}},
+{"objectID":"b739fcf7819dd6f055ebe3bcdb161ff839067dc172235f9d0072f9cf79d7bcd3","title":"Provider Credential Reference","url":"/docs/features/per-request-credentials#provider-credential-reference","content":"All fields are optional — omit any field you want to fall through to a lower-precedence level.\n\n| Provider             | Key                 | Fields                                                                                                        |\n| -------------------- | ------------------- | ------------------------------------------------------------------------------------------------------------- |\n| OpenAI               | openai            | apiKey, baseURL                                                                                           |\n| Anthropic            | anthropic         | apiKey, oauthToken, baseURL                                                                             |\n| Google AI Studio     | googleAiStudio    | apiKey, baseURL                                                                                           |\n| Google Vertex AI     | vertex            | projectId, location, apiKey (Express Mode), serviceAccountKey, clientEmail, privateKey, baseURL |\n| Amazon Bedrock       | bedrock           | accessKeyId, secretAccessKey, sessionToken, region                                                    |\n| Amazon SageMaker     | sagemaker         | accessKeyId, secretAccessKey, sessionToken, region, endpoint                                        |\n| Azure OpenAI         | azure             | apiKey, resourceName, deploymentName, apiVersion                                                      |\n| Mistral              | mistral           | apiKey                                                                                                      |\n| Hugging Face         | huggingFace       | apiKey, baseURL                                                                                           |\n| OpenRouter           | openrouter        | apiKey, baseURL                                                                                           |\n| LiteLLM              | litellm           | apiKey, baseURL","hierarchy":{"lvl0":"Features","lvl1":"Per-Request Credentials","lvl2":"Provider Credential Reference","lvl3":""}},
 {"objectID":"575e2e50c5fd00e53fce66deee556264f3df8962e2188ac446e6e7d4ad29b2ed","title":"OpenAI","url":"/docs/features/per-request-credentials#openai","content":"Custom base URL for OpenAI-compatible proxies:","hierarchy":{"lvl0":"Features","lvl1":"Per-Request Credentials","lvl2":"OpenAI","lvl3":""}},
 {"objectID":"499e7c1d012d337ad54e06db35770d5b3471afb0f9e3f471564642e8ba7e5c6b","title":"Anthropic","url":"/docs/features/per-request-credentials#anthropic","content":"OAuth token (Anthropic Claude subscription):\n\nA gateway or proxy in front of Anthropic (baseURL wins over ANTHROPIC_BASE_URL, with or without a trailing /v1; a gateway that does its own auth accepts any apiKey):","hierarchy":{"lvl0":"Features","lvl1":"Per-Request Credentials","lvl2":"Anthropic","lvl3":""}},
 {"objectID":"3f5ea871d2750f36fca227c33f109de53f5d0bc4c184ea956d0b8c3736babe29","title":"Vertex AI — Express Mode","url":"/docs/features/per-request-credentials#vertex-ai-express-mode","content":"Express Mode uses a simple API key instead of service-account credentials, making it suitable for per-request BYOK flows:\n\nFull service-account credentials (server-side only — keep private keys out of client code):","hierarchy":{"lvl0":"Features","lvl1":"Per-Request Credentials","lvl2":"Vertex AI — Express Mode","lvl3":""}},

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@juspay/neurolink",
-  "version": "12.49.4",
+  "version": "12.49.6",
   "packageManager": "pnpm@10.15.1",
   "description": "The pipe layer of an AI nervous system: one interface connecting provider neurons to your application, across three inference types — generate, stream and decide. `decide` returns typed, calibrated judgements from a non-generative model (~400ms, ~$0.00002/call) for routing, tool selection and context budgeting. MCP-native (4 transports), voice TTS/STT/realtime, RAG, agents, memory, compaction, 9 observability exporters. OpenAI · Anthropic · Gemini · Bedrock · Azure · Ollama · TypeSafe Jev and more.",
   "author": {