pi-freeflow 1.4.3 → 1.4.5

package/README.md

@@ -1,6 +1,6 @@
 # pi-freeflow 🌊
 
-> **23 free models. Up to 1M context. Zero API keys. Infinite scale via your own relay pool.**
+> **21 free models. Up to 1M context. Zero API keys. Infinite scale via your own relay pool.**
 
 Thin by design: model list + dumb relay + log. Host `pi-ai` owns thinking, normalization & provider magic. We just make it free, fast, and unbreakable.
 
@@ -18,7 +18,7 @@ Join devs bypassing rate limits with their own relay pools. BYO, add as many as
 
 | Feature | Description | Value | Cost |
 | :--- | :--- | :--- | :--- |
-| **23 Curated Free Models** | 9 OpenCode Zen + 14 KiloCode Gateway models, up to 1M context & 512K output | Ceiling Unlocked | **$0** |
+| **21 Curated Free Models** | 7 OpenCode Zen + 14 KiloCode Gateway models, up to 1M context & 512K output | Ceiling Unlocked | **$0** |
 | **BYO Relay Pool** | Round-robin load balancing across your Cloudflare Workers & Vercel Edges | Zero Rate Limits | **$0** (your free tiers) |
 | **Adaptive Health & Error Detection** | Auto-cooldown on 429 rate limits, 504 timeouts, and socket drops | 0ms Wasted Latency | **$0** |
 | **Stream Truncation Resilience** | Stateful SSE terminal tracking (`response.failed` / `response.incomplete` injection) | Zero Host Crashes | **$0** |
@@ -32,21 +32,19 @@ Philosophy: **Thin by design.** We only ship model list + relay proxy + log. Hos
 
 ---
 
-### 23 Curated Models, One Command
+### 21 Curated Models, One Command
 
 ```bash
 /model → freeflow → pick
 ```
 
-#### OpenCode Zen (9 Models), Responses & Chat API
+#### OpenCode Zen (7 Models), Responses & Chat API
 Optimized for deep reasoning, long-horizon coding & autonomous agentic workflows.
 
 | Model ID | Creator / Lab | Context | Max Output | Thinking | Vision |
 | :--- | :--- | :--- | :--- | :--- | :--- |
 | `muse-spark-1.2-contributor-free` | Meta Superintelligence Labs | **1M** (1.048.576) | **131K** (131.072) | `minimal / low / medium / high / xhigh / max` | ✅ |
-| `x-preview-f-free` | Ox Alpha | **1M** (1.048.576) | **131K** (131.072) | `low / high / max` | ✅ |
 | `mimo-v2.5-free` | Xiaomi MiMo | **1M** (1.048.576) | **131K** (131.072) | `low / medium / high` | ✅ |
-| `deepseek-v4-flash-free` | DeepSeek | **1M** (1.000.000) | **384K** (384.000) | `low / high / max` | ❌ |
 | `laguna-s-2.1-free` | Poolside | **1M** (1.048.576) | **131K** (131.072) | `low / high / max` | ❌ |
 | `nemotron-3.5-lightning-free` | NVIDIA | **1M** (1.000.000) | **262K** (262.144) | `low / high / max` | ❌ |
 | `nemotron-3-ultra-free` | NVIDIA | **1M** (1.000.000) | **128K** (128.000) | `low / high / max` | ❌ |
@@ -107,7 +105,7 @@ Manage your relay pool directly from the OMP / Pi terminal:
 /freeflow label <index|url> <name># Assign a friendly label to a relay
 /freeflow remove <index|url|label># Remove a relay from the pool
 /freeflow on | off | auto         # Toggle relay mode (auto = enabled for freeflow)
-/freeflow deploy vercel            # Guided Vercel Edge relay deploy (also: deploy cloudflare, deploy deno)
+/freeflow deploy <platform>       # Guided relay deploy: vercel|cloudflare|deno — token in-memory, auto-adds (Vercel 1M/mo recommended)
 /freeflow logs [lines]            # Inspect recent proxy logs
 /freeflow debug on | off          # Toggle full HTTP lifecycle debug logging
 ```
@@ -148,9 +146,11 @@ Default ships direct. Add relays via `/freeflow add <url> [label]`.
 
 **Zero setup?** Run `/freeflow deploy cloudflare` (or `deno`, `vercel`), paste your platform token once, and the relay is created and activated for you. Manual snippets below.
 
-**Option A: Cloudflare Workers (100k req/day, no 25s timeout)**
-1. Go to `dash.cloudflare.com` → Workers → Create → Deploy → Edit code
-2. Paste this Worker relay snippet:
+**Option A: Cloudflare Workers (100k req/day, no 25s timeout) — Auto Deploy**
+```bash
+/freeflow deploy cloudflare  # prompts token in-memory, auto-adds to pool
+```
+*Manual fallback:* `dash.cloudflare.com` → Workers → Create → Deploy → Edit code → paste snippet below → Deploy → `/freeflow add https://your.workers.dev cf-worker-1`
 
 ```js
 // Only the 2 upstreams pi-freeflow talks to. Anything else = open proxy abuse.
@@ -170,20 +170,18 @@ export default {
   },
 };
 ```
-3. Deploy and add it:
 
+**Option B: Vercel Edge Relay (1M req/mo) — Auto Deploy**
 ```bash
-omp → /freeflow add https://your.workers.dev cf-worker-1
+/freeflow deploy vercel  # prompts token in-memory, auto-adds to pool
+# or shorthand: /freeflow deploy
 ```
 
-**Option B: Vercel Edge Relay (1M req/mo), Auto Deploy**
+**Option C: Deno Deploy (100k req/day) — Auto Deploy**
 ```bash
-/freeflow deploy  # prompts token in-memory, auto-adds to pool
+/freeflow deploy deno  # prompts token in-memory, auto-adds to pool
 ```
-
-**Option C: Deno Deploy (100k req/day)**
-1. Go to `dash.deno.com` → New Project → Playground
-2. Paste this relay and click **Deploy**:
+*Manual fallback:* `dash.deno.com` → New Project → Playground → paste snippet below → Deploy → `/freeflow add https://your-project.deno.dev deno-relay-1`
 
 ```ts
 const ALLOWED_TARGETS = ["https://opencode.ai", "https://api.kilo.ai"];
@@ -201,12 +199,6 @@ Deno.serve(async (req) => {
 });
 ```
 
-3. Add it:
-
-```bash
-/freeflow add https://your-project.deno.dev deno-relay-1
-```
-
 **Verify your pool:**
 ```bash
 /freeflow status        # relay-A 1/N (ON) → candidates:N
@@ -285,7 +277,7 @@ pnpm smoke       # verifies extensions/index.ts loads without crashing
 ```
 src/
 ├── index.ts          # extension entry, lifecycle hooks
-├── models.ts         # 23-model catalog definitions
+├── models.ts         # 21-model catalog definitions
 ├── catalog.ts        # model catalog cache (24h disk)
 ├── proxy.ts          # local proxy server (127.0.0.1:18080)
 ├── relay.ts          # relay selection & round-robin

package/package.json

@@ -1,7 +1,7 @@
 {
 	"name": "pi-freeflow",
 	"type": "module",
-	"version": "1.4.3",
+	"version": "1.4.5",
 	"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

@@ -31,9 +31,14 @@ import type {
 	Upstream,
 } from "./types.ts";
 
+/**
+ * Pruned model IDs that must never re-enter the catalog via disk cache or upstream merge.
+ */
+const DEAD_MODEL_IDS = new Set<string>(["deepseek-v4-flash-free", "x-preview-f-free"]);
+
 /**
  * In-memory cache of currently active/available free models.
- * Initialized with all 23 verified models for 0ms instant availability.
+ * Initialized with all 21 verified models for 0ms instant availability.
  */
 let aliveCatalog: RegisteredModel[] = ALL_MODELS.map((m) => ({
 	...m,
@@ -58,14 +63,17 @@ export function setAliveCatalog(catalog: RegisteredModel[]): void {
  * Fresh entries win on id collision; unknown fresh ids are appended after the base.
  * Guards the background refresh against a partial upstream cache removing
  * verified static models from provider registration.
+ * Filters pruned dead IDs so they never re-enter via fresh upstream data.
  */
 export function mergeCatalog(
 	base: RegisteredModel[],
 	fresh: RegisteredModel[],
 ): RegisteredModel[] {
+	const filteredFresh = fresh.filter((m) => !DEAD_MODEL_IDS.has(m.id));
 	const byId = new Map(base.map((m) => [m.id, m]));
-	for (const m of fresh) byId.set(m.id, m);
-	return [...byId.values()];
+	for (const m of filteredFresh) byId.set(m.id, m);
+	// Ensure no dead IDs survive even if base was stale
+	return [...byId.values()].filter((m) => !DEAD_MODEL_IDS.has(m.id));
 }
 
 /**
@@ -122,7 +130,6 @@ export function enrichModelDef(raw: RawModelItem, source: Upstream): RegisteredM
 		idLower.includes("r1") ||
 		idLower.includes("o1") ||
 		idLower.includes("think") ||
-		idLower.includes("alpha") ||
 		idLower.includes("spark");
 
 	let contextWindow =
@@ -162,6 +169,7 @@ export function enrichModelDef(raw: RawModelItem, source: Upstream): RegisteredM
 
 /**
  * Read cached catalog data from disk if valid and unexpired.
+ * Filters out pruned dead IDs so stale disk entries never repopulate the catalog.
  */
 export function readCatalogCache(): CatalogCacheData | null {
 	try {
@@ -170,6 +178,9 @@ export function readCatalogCache(): CatalogCacheData | null {
 		}
 		const raw = fs.readFileSync(CATALOG_CACHE_FILE, "utf8");
 		const data = JSON.parse(raw) as CatalogCacheData;
+		if (Array.isArray(data.models)) {
+			data.models = data.models.filter((m) => !DEAD_MODEL_IDS.has(m.id));
+		}
 		if (Date.now() - data.timestamp < CATALOG_CACHE_TTL_MS) {
 			return data;
 		}
@@ -203,20 +214,37 @@ export function writeCatalogCache(data: CatalogCacheData): void {
 export async function refreshCatalog(force = false): Promise<RegisteredModel[]> {
 	// Thin provider: no live fetch — subagents must not hit upstream directly
 	// (proxy-only). Host Pi/OMP owns dynamic discovery via fetchDynamicModels (24h).
-	// We only serve disk cache if fresh, otherwise static 23-model aliveCatalog.
+	// We only serve disk cache if fresh, otherwise static 21-model aliveCatalog.
 	const disk = readCatalogCache();
 	if (disk && Array.isArray(disk.models) && disk.models.length > 0) {
 		const age = Date.now() - (disk.timestamp ?? 0);
 		if (!force && age < CATALOG_CACHE_TTL_MS) {
-			aliveCatalog = disk.models;
+			aliveCatalog = disk.models.filter((m) => !DEAD_MODEL_IDS.has(m.id));
 			return aliveCatalog;
 		}
-		// Stale cache still better than empty — return it without network
-		if (disk.models.length >= 23) {
-			aliveCatalog = disk.models;
+		// Stale cache still better than empty — return it without network (filtered)
+		const filtered = disk.models.filter((m) => !DEAD_MODEL_IDS.has(m.id));
+		if (filtered.length >= 21) {
+			aliveCatalog = filtered;
 			return aliveCatalog;
 		}
 	}
-	// No valid cache — return in-memory static 23 (host will refresh if needed)
+	// No valid fresh cache — try stale disk cache directly (readCatalogCache returns null when expired)
+	try {
+		if (fs.existsSync(CATALOG_CACHE_FILE)) {
+			const raw = fs.readFileSync(CATALOG_CACHE_FILE, "utf8");
+			const stale = JSON.parse(raw) as CatalogCacheData;
+			if (Array.isArray(stale.models) && stale.models.length > 0) {
+				const filtered = stale.models.filter((m) => !DEAD_MODEL_IDS.has(m.id));
+				if (filtered.length >= 21) {
+					aliveCatalog = filtered;
+					return aliveCatalog;
+				}
+			}
+		}
+	} catch (err) {
+		logDebug("Failed reading stale catalog cache", { error: String(err) });
+	}
+	// No valid cache — return in-memory static 21 (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 23 free models (9 OpenCode Zen + 14 KiloCode Gateway) with:
+ * Provides access to 21 free models (7 OpenCode Zen + 14 KiloCode Gateway) with:
  * - Single-port daemon reuse on 18080 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 23 verified free models:
- * - 9 OpenCode Zen models (1 Responses API + 8 Chat Completions)
+ * Defines the 21 verified free models:
+ * - 7 OpenCode Zen models (1 Responses API + 6 Chat Completions)
  * - 14 KiloCode Keyless Gateway models (10 OpenRouter format + 4 Standard format)
  */
 
@@ -13,40 +13,6 @@ import type { ModelDef, Upstream } from "./types.ts";
  * Endpoint: https://opencode.ai/zen/v1
  */
 export const OPENCODE_MODELS: ModelDef[] = [
-	{
-		id: "deepseek-v4-flash-free",
-		name: "DeepSeek V4 Flash (1M)",
-		reasoning: true,
-		contextWindow: 1_000_000,
-		maxTokens: 384_000,
-		input: ["text"],
-		thinkingLevelMap: {
-			off: null,
-			minimal: "low",
-			low: "low",
-			medium: "high",
-			high: "high",
-			xhigh: "max",
-			max: "max",
-		},
-	},
-	{
-		id: "x-preview-f-free",
-		name: "Ox Alpha (1M)",
-		reasoning: true,
-		contextWindow: 1_048_576,
-		maxTokens: 131_072,
-		input: ["text", "image"],
-		thinkingLevelMap: {
-			off: null,
-			minimal: "low",
-			low: "low",
-			medium: "high",
-			high: "high",
-			xhigh: "max",
-			max: "max",
-		},
-	},
 	{
 		id: "muse-spark-1.2-contributor-free",
 		name: "Muse Spark 1.2 (1M)",
@@ -305,37 +271,27 @@ export const KILO_MODELS: ModelDef[] = [
 
 /**
  * Model ID Aliases — maps user-friendly / slash-free CLI IDs to canonical upstream model IDs.
+ * Single clean alias per model (no :free duplicates). Wrong cross-lab aliases removed:
+ * - Muse Spark 1.2 = Meta Superintelligence Labs (not Anthropic Claude) → removed claude-sonnet aliases
+ * - Laguna S 2.1 = Poolside (not MiniMax) → removed minimax-m2.1 alias
+ * - Hy3 = Tencent Hunyuan (not Alibaba Qwen) → removed qwen3-coder alias
  */
 export const MODEL_ALIASES: Record<string, string> = {
-	// Kilo Gateway slash-free & colon-free CLI aliases
+	// Kilo Gateway — slash-free & colon-free clean aliases (one per model)
 	"dots-3-note-preview": "dots-studio/dots-3-note-preview:free",
-	"dots-3-note-preview:free": "dots-studio/dots-3-note-preview:free",
 	"step-3.7-flash": "stepfun/step-3.7-flash:free",
-	"step-3.7-flash:free": "stepfun/step-3.7-flash:free",
 	"nemotron-3-nano-omni": "nvidia/nemotron-3-nano-omni-30b-a3b-reasoning:free",
-	"nemotron-3-nano-omni:free": "nvidia/nemotron-3-nano-omni-30b-a3b-reasoning:free",
 	"nemotron-3-ultra-550b": "nvidia/nemotron-3-ultra-550b-a55b:free",
-	"nemotron-3-ultra-550b:free": "nvidia/nemotron-3-ultra-550b-a55b:free",
 	"nemotron-3-super": "nvidia/nemotron-3-super-120b-a12b:free",
-	"nemotron-3-super:free": "nvidia/nemotron-3-super-120b-a12b:free",
-	"hy3:free": "tencent/hy3:free",
 	"north-mini-code": "cohere/north-mini-code:free",
-	"north-mini-code:free": "cohere/north-mini-code:free",
-	"laguna-s-2.1:free": "poolside/laguna-s-2.1:free",
-	"laguna-xs-2.1:free": "poolside/laguna-xs-2.1:free",
 	"lfm-2.5": "liquid/lfm-2.5-2.6b:free",
-	"lfm-2.5:free": "liquid/lfm-2.5-2.6b:free",
 	"content-safety": "nvidia/nemotron-3.5-content-safety:free",
-	"content-safety:free": "nvidia/nemotron-3.5-content-safety:free",
+	// provider-prefixed short aliases (slash-normalized)
+	"hy3:free": "tencent/hy3:free",
+	"laguna-s-2.1:free": "poolside/laguna-s-2.1:free",
+	"laguna-xs-2.1:free": "poolside/laguna-xs-2.1:free",
 	"kilo-auto": "kilo-auto/free",
 	"openrouter": "openrouter/free",
-
-	// OpenCode Zen aliases
-	"claude-sonnet-4.5-free": "muse-spark-1.2-contributor-free",
-	"claude-sonnet-4.5-contributor-free": "muse-spark-1.2-contributor-free",
-	"grok-code-fast-1-preview-f-free": "x-preview-f-free",
-	"minimax-m2.1-free": "laguna-s-2.1-free",
-	"qwen3-coder-480b-free": "hy3-free",
 };
 
 /**
@@ -357,7 +313,7 @@ export const KILO_MODEL_IDS = new Set<string>([
 ]);
 
 /**
- * Combined list of all 23 static free models (canonical)
+ * Combined list of all 21 static free models (canonical)
  */
 export const ALL_MODELS: ModelDef[] = [...OPENCODE_MODELS, ...KILO_MODELS];
 

package/src/proxy.ts

@@ -88,21 +88,11 @@ export function sanitizeHeaders(
 }
 
 /**
- * Clamps reasoning_effort for upstream models with strict non-standard enums
- * (e.g. OpenCode x-preview strictly requires 'low', 'high', or 'max' and rejects 'medium' with 400).
+ * Reasoning normalization is owned by host pi-ai; proxy passes through reasoning fields unchanged.
+ * Kept as no-op for compatibility — no per-model clamping.
  */
-function sanitizeReasoningForModel(bodyObj: Record<string, unknown>): void {
-	const model = String(bodyObj.model || "").toLowerCase();
-	if (model.includes("x-preview")) {
-		const effort = String(bodyObj.reasoning_effort || "").toLowerCase();
-		if (effort === "medium") {
-			bodyObj.reasoning_effort = "high";
-		} else if (effort === "minimal") {
-			bodyObj.reasoning_effort = "low";
-		} else if (!effort || effort === "off" || effort === "none") {
-			bodyObj.reasoning_effort = "low";
-		}
-	}
+function sanitizeReasoningForModel(_bodyObj: Record<string, unknown>): void {
+	// no-op
 }
 /**
  * Probe whether an existing pi-freeflow proxy daemon is running and responsive on a given port.