MythEclipse/GMW
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

chore: migrate AI LLM router from 9router to omniroute

Browse Source 
Switch GMW's AI LLM base URL from 9router (https://9router.asepharyana.my.id/v1)
to omniroute on imrnes (http://100.121.180.82:20128/api/v1).

- Update default AI_LLM_BASE_URL in discord-gateway + backend config schemas
- Update .env.example documentation
- Update all 9router references in comments/docs/tests to omniroute
- Production BWS secret gmw_ai_llm_base_url already updated

Omniroute uses /api/v1 prefix (not /v1 like 9router), so the base URL
now correctly points at the right API path for the OpenAI SDK.
This commit is contained in:
asepharyana
2026-08-28 20:18:48 +07:00
parent 14bd20f072
commit ffbe9959ab
9 changed files with 16 additions and 16 deletions
Download Patch File Download Diff File Expand all files Collapse all files
.env.example
+1 -1
View File
@@ -85,7 +85,7 @@ BACKLOG_SYNC_BATCH_SIZE=100 # Messages per backlog batch, max 100 (d
# === AI Analysis ===
AI_ANALYSIS_ENABLED=false # Enable AI content moderation (default: false)
AI_LLM_API_KEY= # REQUIRED if AI_ANALYSIS_ENABLED=true. LLM API key
AI_LLM_BASE_URL=https://9router.asepharyana.my.id/v1 # LLM API base URL (9router — OpenAI-compatible router, replaces omniroute)
AI_LLM_BASE_URL=http://100.121.180.82:20128/api/v1 # LLM API base URL (omniroute — OpenAI-compatible router on imrnes, Tailscale 100.121.180.82)
AI_LLM_MODEL=text # LLM text model name (default: text)
# AI_LLM_VISION_MODEL= # Vision model for image analysis (falls back to AI_LLM_MODEL)
# AI_LLM_EMBEDDING_MODEL= # Embedding model for semantic moderation cache (optional; enables near-duplicate text reuse to save LLM calls)
services/backend/src/modules/chatbot/chatbot.service.ts
+4 -4
View File
@@ -158,7 +158,7 @@ Gaya ngobrol:
tool_choice: "auto",
max_tokens: 600,
temperature: 0.4,
// Non-streaming: request a single complete response. 9router may
// Non-streaming: request a single complete response. omniroute may
// still emit SSE even with stream:false, so the parser below
// handle both raw-JSON and SSE bodies.
stream: false,
@@ -176,7 +176,7 @@ Gaya ngobrol:
},
);
// Parse the body into content + tool_calls. 9router may return either
// Parse the body into content + tool_calls. omniroute may return either
// a single JSON object (stream:false honored) or SSE text (stream
// implied) — parseResponse handles both.
const { content, toolCalls } = this.parseResponse(
@@ -252,7 +252,7 @@ Gaya ngobrol:
/**
* Parse an LLM HTTP body into content + tool_calls. Handles both shapes
* 9router can return: a single JSON object (stream:false honored) or SSE
* omniroute can return: a single JSON object (stream:false honored) or SSE
* text (stream implied). For SSE we delegate to parseSse.
*/
private parseResponse(body: string): {
@@ -306,7 +306,7 @@ Gaya ngobrol:
/**
* Parse an SSE stream body into accumulated content + any tool_calls.
* 9router (and most OpenAI-compatible routers) emit `data: {json}` lines
* omniroute (and most OpenAI-compatible routers) emit `data: {json}` lines
* even when stream is only implied; we must collect deltas manually.
*/
private parseSse(body: string): {
services/backend/src/shared/config/index.ts
+1 -1
View File
@@ -131,7 +131,7 @@ export const configSchema = z
AI_LLM_BASE_URL: z
.string()
.url()
.default("https://9router.asepharyana.my.id/v1"),
.default("http://100.121.180.82:20128/api/v1"),
AI_LLM_MODEL: z.string().default("text"),
AI_LLM_VISION_MODEL: z.string().optional(),
AI_LLM_EMBEDDING_MODEL: z.string().optional(),
services/discord-gateway/ARCHITECTURE.md
+1 -1
View File
@@ -139,5 +139,5 @@ pipeline gauges — `ai_analysis_queued_conversations`,
numeric snowflake IDs never trigger false positives.
- **Semantic cache is batched** (one embed call + one Qdrant batch search),
not N sequential round-trips. `ensureQdrantCollection` is memoized.
- **Streaming is mandatory** against the 9router base URL (non-stream waits for
- **Streaming is mandatory** against the omniroute base URL (non-stream waits for
the full body and times out). `llmClient` aggregates SSE chunks.
services/discord-gateway/src/modules/ai-moderation/llmCaller.ts
+1 -1
View File
@@ -85,7 +85,7 @@ export async function callModerationLLM(
jsonResponse: { type: "json_object" },
retries: 0,
signal,
// Router (9router / formerly omniroute) always streams SSE even when the
// Router (omniroute) always streams SSE even when the
// request omits `stream`. In non-stream mode the OpenAI SDK waits
// for the FULL body before parsing, so slow/long upstream streams
// hit the 30s/60s timeout and abort mid-generation. Streaming mode
services/discord-gateway/src/modules/ai-moderation/llmClient.ts
+2 -2
View File
@@ -94,7 +94,7 @@ type LLMResponseChunk = {
* `delta.content`; falls back to reasoning fields so reasoning-only models
* still produce usable aggregated text. Providers differ in the field name:
* - DeepSeek-style / Cloudflare gemma → `delta.reasoning_content`
* - mimo (via 9router) streams reasoning in `delta.reasoning` +
* - mimo (via omniroute) streams reasoning in `delta.reasoning` +
* `delta.reasoning_details[].text` (content:"") — without these fallbacks
* vision aggregation came back empty ("Vision API null response").
* Exported for unit tests.
@@ -225,7 +225,7 @@ export function buildLlmParams(
reasoning: { enabled: false },
// vLLM / Qwen / litellm
chat_template_kwargs: { enable_thinking: false },
// Anthropic / Claude-format (9router exposes thinkingFormat
// Anthropic / Claude-format (omniroute exposes thinkingFormat
// "claude-adaptive" / "claude-budget" on its reasoning models)
thinking: { type: "disabled" },
} as Record<string, unknown>);
services/discord-gateway/src/shared/config/index.ts
+1 -1
View File
@@ -149,7 +149,7 @@ export const configSchema = z
AI_LLM_BASE_URL: z
.string()
.url()
.default("https://9router.asepharyana.my.id/v1"),
.default("http://100.121.180.82:20128/api/v1"),
AI_LLM_MODEL: z.string().default("text"),
// Vision uses the SAME router/base URL as text moderation
// (AI_LLM_BASE_URL) but a different model alias. The dedicated NVIDIA
services/discord-gateway/tests/llmChunkExtraction.test.ts
+4 -4
View File
@@ -1,7 +1,7 @@
// ═══════════════════════════════════════════════════════════════════════════
// llmClient chunk extraction — reasoning_content fallback (pure, no network)
// ═══════════════════════════════════════════════════════════════════════════
// Regression: 9router "multimodal" combo routed to cloudflare gemma-4-26b
// Regression: omniroute "multimodal" combo routed to cloudflare gemma-4-26b
// which streams ALL output in delta.reasoning_content with content:"" — the
// old extractor returned empty text → llmVision reported "Vision API null
// response" → every image moderation batch fell back to text-only analysis
@@ -19,7 +19,7 @@ describe("extractChunkText — streaming chunk text extraction", () => {
});
it("falls back to delta.reasoning_content when content is empty — reasoning-only models (cloudflare gemma)", () => {
// Exact shape seen from 9router → cloudflare-ai/@cf/google/gemma-4-26b:
// Exact shape seen from omniroute → cloudflare-ai/@cf/google/gemma-4-26b:
// {"choices":[{"delta":{"content":"","reasoning_content":"Task","role":"assistant"},"finish_reason":null,...}]}
expect(
extractChunkText({
@@ -33,8 +33,8 @@ describe("extractChunkText — streaming chunk text extraction", () => {
).toBe("Task");
});
it('falls back to delta.reasoning — mimo via 9router streams reasoning there with content:""', () => {
// Exact shape seen from 9router → mimo-v2.5-free (2026-08-11):
it('falls back to delta.reasoning — mimo via omniroute streams reasoning there with content:""', () => {
// Exact shape seen from omniroute → mimo-v2.5-free (2026-08-11):
// {"choices":[{"delta":{"content":"","reasoning":"The user wants a","role":"assistant"},"finish_reason":null,...}]}
expect(
extractChunkText({
services/discord-gateway/tests/visionNoImageSeen.test.ts
+1 -1
View File
@@ -6,7 +6,7 @@
// a VALID vision_llm result. Every later analysis of the same image (same
// hash / phash) then hit the poisoned cache and the moderation LLM wrote
// "lampiran yang gagal terbaca" — image analysis seemed permanently broken
// even though 9router was responding fine.
// even though omniroute was responding fine.
import { describe, expect, it } from "vitest";
import { isNoImageSeenText } from "../src/modules/ai-moderation/visionAnalyzer.js";