pi-freeflow 1.15.1 → 1.16.0

package/CHANGELOG.md

@@ -1,5 +1,18 @@
 # Changelog
 
+## 1.16.0
+
+### Minor Changes
+
+- New free model: Union Alpha Free joins the picker. It answers through a different upstream endpoint than the other free models, handled automatically — 262K context, image input, tool calling, and zero cost like the rest of the catalog.
+
+## 1.15.2
+
+### Patch Changes
+
+- Relay traffic is now spent only on the free models the pool exists for: requests for other models go straight upstream even with the relay pool on, cutting relay bandwidth. Applies in automatic and always-on modes; relay-off behavior is unchanged.
+- The disabled-deployment rollover now triggers only on genuine hosting verdicts, so ordinary payment and quota refusals always surface immediately without cooling a healthy relay. Relay-list imports also no longer show a negative removal count when the file holds more relays than the current pool.
+
 ## 1.15.1
 
 ### Patch Changes

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "pi-freeflow",
   "type": "module",
-  "version": "1.15.1",
+  "version": "1.16.0",
   "description": "Thin provider for OMP/Pi — model list + dumb relay proxy + log; host pi-ai owns thinking/normalization",
   "main": "extensions/index.ts",
   "types": "src/index.ts",

package/src/catalog.ts

@@ -77,7 +77,7 @@ export function sanitizeCatalogModels(models: RegisteredModel[]): RegisteredMode
 }
 /**
  * In-memory cache of currently active/available free models.
- * Initialized with all 26 verified models for 0ms instant availability.
+ * Initialized with all 27 verified models for 0ms instant availability.
  */
 let aliveCatalog: RegisteredModel[] = ALL_MODELS.map((m) => ({
 	...m,
@@ -400,6 +400,6 @@ export async function refreshCatalog(force = false): Promise<RegisteredModel[]>
 	} catch (err) {
 		logDebug("Failed reading stale catalog cache", { error: String(err) });
 	}
-	// No valid cache — return in-memory static 26 (host will refresh if needed)
+	// No valid cache — return in-memory static 27 (host will refresh if needed)
 	return aliveCatalog;
 }

package/src/index.ts

@@ -1,7 +1,7 @@
 /**
  * pi-freeflow — Modular, high-resiliency LLM extension for Pi & Oh My Pi (OMP)
  *
- * Provides access to 26 free models (7 OpenCode Zen + 19 KiloCode Gateway) with:
+ * Provides access to 27 free models (8 OpenCode Zen + 19 KiloCode Gateway) with:
  * - Single-port daemon reuse on 28180 across concurrent subagents
  * - Multi-cloud rolling egress relays (Vercel Edge, Cloudflare, Deno)
  * - 0ms instant startup with verified static catalog and background live health checks

package/src/models.ts

@@ -1,8 +1,8 @@
 /**
  * Static model definitions and upstream routing catalogs for pi-freeflow
  *
- * Defines the 26 verified free models:
- * - 7 OpenCode Zen models (2 Responses API + 5 Chat Completions)
+ * Defines the 27 verified free models:
+ * - 8 OpenCode Zen models (2 Responses API + 5 Chat Completions + 1 Anthropic Messages)
  * - 19 KiloCode Keyless Gateway models (18 OpenRouter format + 1 Standard format)
  */
 
@@ -127,6 +127,19 @@ export const OPENCODE_MODELS: ModelDef[] = [
 			xhigh: "xhigh",
 			max: null,},
 	},
+	{
+		id: "union-alpha",
+		name: "Union Alpha Free",
+		reasoning: true,
+		contextWindow: 262_144,
+		maxTokens: 131_072,
+		api: "anthropic-messages",
+		input: ["text", "image"],
+		// No thinkingLevelMap: the effort wire values for this endpoint are
+		// unknown, so the host sends no effort param. The model answers
+		// anyway (proven live 2026-09-17: POST /zen/v1/messages streams real
+		// tokens at zero cost). Followup: probe effort values, then add a map.
+	},
 ];
 
 /**
@@ -389,7 +402,7 @@ export const KILO_MODEL_IDS = new Set<string>([
 ]);
 
 /**
- * Combined list of all 26 static free models (canonical)
+ * Combined list of all 27 static free models (canonical)
  */
 export const ALL_MODELS: ModelDef[] = [...OPENCODE_MODELS, ...KILO_MODELS];
 

package/src/proxy.ts

@@ -89,6 +89,21 @@ function withRateLimitHint(status: number, data: string): string {
 	return data;
 }
 
+/**
+ * Relay eligibility: only free-catalog models ride relay egress. Returns true
+ * when the model is a non-empty string present in MODEL_MAP or KILO_MODEL_IDS
+ * (Kilo traffic leaves via its own branch regardless), and when the model is
+ * missing/empty (e.g. GET /v1/models carries no body — preserve routing).
+ * Anything else (unknown/paid model ids, non-strings) goes direct upstream.
+ */
+export function isRelayEligibleModel(model: unknown): boolean {
+	if (model === undefined || model === null) return true;
+	if (typeof model !== "string") return false;
+	if (model.trim() === "") return true;
+	const canonical = resolveCanonicalModelId(model.trim());
+	return MODEL_MAP.has(canonical) || KILO_MODEL_IDS.has(canonical);
+}
+
 /**
  * Extract client IP address from incoming HTTP request.
  */
@@ -692,12 +707,18 @@ export function startProxy(
 						res.end(data);
 					}
 				} else {
-					// OpenCode routing — relay when enabled, else direct upstream
+					// OpenCode routing — relay egress is for free-catalog models
+					// only; unknown/paid models go direct upstream.
 					const relayState = getActiveRelayState();
-					const shouldUseRelay =
+					const relayPoolOn =
 						relayState.mode !== "off" &&
 						relayState.enabled !== false &&
 						Boolean(relayState.url || (relayState.relays && relayState.relays.length > 0));
+					const relayEligible = isRelayEligibleModel(parsedBody?.model);
+					const shouldUseRelay = relayPoolOn && relayEligible;
+					if (relayPoolOn && !relayEligible) {
+						log("info", `relay bypass for non-catalog model ${String(parsedBody?.model ?? "?")} — routing direct upstream`, { model: parsedBody?.model }, reqId);
+					}
 					if (shouldUseRelay) {
 						const fullUrl = `${UPSTREAM_OPENCODE}${req.url ?? "/"}`;
 						const activeHost = relayState.url