Skip to content

Never stop coding. Free MIT AI gateway: one endpoint, 375 providers (150+ free), 1,400+ models Kimi, Claude, GPT, Gemini, GLM, DeepSeek, MiniMax. Works with Claude Code, Codex, Cursor, OpenCode, Cline & Copilot. Quota-aware auto-fallback, RTK+Caveman compression saves 15-95% tokens, MCP/A2A, Desktop/PWA. Built by 900+ contributors

fix(nous-research): correct baseUrl to include /chat/completions (#2826) (#2835)

Diego Rodrigues de Sa e Souza283c05d3d082
diff --git a/CHANGELOG.md b/CHANGELOG.md
index 40c9612d9b..3c365fdbb4 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -14,6 +14,7 @@
### 🔧 Bug Fixes
+- **nous-research:** append /chat/completions to provider baseUrl so DefaultExecutor's default URL builder hits the correct endpoint instead of returning 404 ([#2826])
- **fix(quota):** honor explicit per-connection `quotaPreflightEnabled: false` even when the provider has global window defaults — adds early-return guard before the AND-of-negations gate in auth.ts ([#2831])
- **api:** include noAuth providers (opencode, etc.) in `/v1/models` active aliases so their models surface without a DB connection row (#2798)
- **opencode-go:** route Qwen3.x via Claude messages format and repair `fixMissingToolResponses` helper for Claude-shape upstreams (#2791 — thanks @jeferssonlemes)
diff --git a/open-sse/config/providerRegistry.ts b/open-sse/config/providerRegistry.ts
index 8841a21266..591a92829c 100644
--- a/open-sse/config/providerRegistry.ts
+++ b/open-sse/config/providerRegistry.ts
@@ -3955,7 +3955,7 @@ export const REGISTRY: Record<string, RegistryEntry> = {
alias: "nous",
format: "openai",
executor: "default",
- baseUrl: "https://inference-api.nousresearch.com/v1",
+ baseUrl: "https://inference-api.nousresearch.com/v1/chat/completions",
authType: "apikey",
authHeader: "bearer",
models: [
diff --git a/tests/unit/executor-nous-research.test.ts b/tests/unit/executor-nous-research.test.ts
new file mode 100644
index 0000000000..a20e9735c0
--- /dev/null
+++ b/tests/unit/executor-nous-research.test.ts
@@ -0,0 +1,36 @@
+/**
+ * Regression test for #2826 — NOUS-RESEARCH provider always returns 404 Not Found
+ *
+ * `nous-research` uses `executor: "default"`. `DefaultExecutor.buildUrl()`'s default
+ * switch case returns `config.baseUrl` verbatim. Without `/chat/completions` appended
+ * every outbound request hits `/v1` and gets a 404.
+ *
+ * Fix: set `baseUrl` in providerRegistry.ts to include `/chat/completions`.
+ */
+import test from "node:test";
+import assert from "node:assert/strict";
+
+import { DefaultExecutor } from "../../open-sse/executors/default.ts";
+
+test("nous-research DefaultExecutor.buildUrl() returns a URL ending with /chat/completions", () => {
+ const executor = new DefaultExecutor("nous-research");
+
+ const url = executor.buildUrl("Hermes-4-405B", true, 0, null);
+
+ assert.match(
+ url,
+ /\/chat\/completions$/,
+ `Expected URL to end with /chat/completions, got: ${url}`
+ );
+});
+
+test("nous-research DefaultExecutor.buildUrl() targets the correct inference endpoint", () => {
+ const executor = new DefaultExecutor("nous-research");
+
+ const url = executor.buildUrl("Hermes-4-70B", false, 0, null);
+
+ assert.equal(
+ url,
+ "https://inference-api.nousresearch.com/v1/chat/completions"
+ );
+});