pi-free 1.0.7 → 1.0.9

package/CHANGELOG.md

@@ -7,6 +7,28 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
+## [1.0.9] - 2026-04-14
+
+### Fixed
+- **Qwen OAuth breaks other OAuth providers** — `modifyModels` receives all models across every registered provider, not just Qwen's. The previous `map()` stamped the Qwen dashscope `baseUrl` onto every model, causing other OAuth providers (Kilo, OpenRouter, etc.) to return 404 after a `/login qwen` flow. Now only models with `provider === PROVIDER_QWEN` are patched; others pass through unchanged.
+
+## [1.0.8] - 2026-04-13
+
+### Added
+- **Modal provider** — Free access to GLM-5.1 FP8 (128k context, 16k max output) during promotional period (free until April 30, 2026)
+  - Requires a free Modal API key (`MODAL_API_KEY` or `modal_api_key` in `~/.pi/free.json`)
+  - Model: `zai-org/GLM-5.1-FP8` — 128k context window, 16k max output tokens
+- **Qwen provider** — Free access to Qwen Coder (1,000 requests/day) via OAuth device flow
+  - Run `/login qwen` to authenticate through Qwen Studio (chat.qwen.ai)
+  - Uses `coder-model` alias (maps to Qwen3.6-Plus on the backend)
+  - 131k context window, 16k max output tokens, zero cost
+
+### Fixed
+- **Qwen OAuth browser launch on Windows** — URLs with `&` query params were truncated by `cmd.exe`'s `&` command separator; switched to `powershell.exe Start-Process` which passes the URL as a literal string
+- **Qwen API endpoint** — Replicates qwen-code's `getCurrentEndpoint()` logic: uses `resource_url` from OAuth token response (`dashscope.aliyuncs.com` for Chinese accounts, `portal.qwen.ai` for international), with fallback to `dashscope.aliyuncs.com/compatible-mode/v1`
+- **Qwen DashScope headers** — Added all headers required by DashScope's OpenAI-compatible API: `X-DashScope-AuthType: qwen-oauth`, `X-DashScope-CacheControl: enable`, `X-DashScope-UserAgent`, `Client-Code: QwenCode`
+- **Qwen modifyModels crash** — `modifyModels` must be synchronous; making it async caused the pi framework to receive a `Promise` instead of a `Model[]`, breaking `ModelRegistry.find()`
+
 ## [1.0.5] - 2025-04-03
 
 ### Fixed

package/README.md

@@ -6,11 +6,11 @@ Free AI model providers for [Pi](https://pi.dev). Access **free models** from mu
 
 ## What does pi-free do
 
-**pi-free is a Pi extension that unlocks free AI models from 8 different providers.**
+**pi-free is a Pi extension that unlocks free AI models from 10 different providers.**
 
 When you install pi-free, it:
 
-1. **Registers 7 AI providers** with Pi's model picker — OpenCode Zen, Kilo, OpenRouter, NVIDIA NIM, Cline, Mistral, and Ollama Cloud
+1. **Registers 9 AI providers** with Pi's model picker — OpenCode Zen, Kilo, OpenRouter, NVIDIA NIM, Cline, Mistral, Ollama Cloud, Qwen, and Modal
 
 2. **Filters to show only free models by default** — You see only the models that cost $0 to use, no API key required for some providers
 
@@ -44,6 +44,8 @@ Free models are shown by default — look for the provider prefixes:
 - `cline/` — Cline models (run `/login cline` to use)
 - `mistral/` — Mistral models (API key required)
 - `ollama/` — Ollama Cloud models (API key required)
+- `qwen/` — Qwen Coder (run `/login qwen` to use, 1,000 free requests/day)
+- `modal/` — GLM-5.1 FP8 via Modal (API key required, free promotional period until April 30, 2026)
 
 ### 3. Toggle between free and paid models
 
@@ -77,7 +79,8 @@ Add your API keys to this file:
   "nvidia_api_key": "nvapi-...",
   "ollama_api_key": "...",
   "fireworks_api_key": "...",
-  "mistral_api_key": "..."
+  "mistral_api_key": "...",
+  "modal_api_key": "sk-modal-..."
 }
 ```
 
@@ -92,8 +95,10 @@ See the [Providers That Need Authentication](#providers-that-need-authentication
 | `/{provider}-toggle` | Switch between free-only and all models for that provider |
 | `/login kilo` | Start OAuth flow for Kilo |
 | `/login cline` | Start OAuth flow for Cline |
+| `/login qwen` | Start OAuth flow for Qwen |
 | `/logout kilo` | Clear Kilo OAuth credentials |
 | `/logout cline` | Clear Cline OAuth credentials |
+| `/logout qwen` | Clear Qwen OAuth credentials |
 
 ---
 
@@ -225,6 +230,29 @@ This command will:
 - Free account required (no credit card)
 - Uses local ports 48801-48811 for OAuth callback
 
+### Modal (GLM-5.1 FP8 — free until April 30, 2026)
+
+Modal hosts GLM-5.1 FP8 with a free promotional period. Get an API key at [modal.com](https://modal.com), then:
+
+**Option A: Environment variable**
+```bash
+export MODAL_API_KEY="sk-modal-..."
+```
+
+**Option B: Config file** (`~/.pi/free.json`)
+```json
+{
+  "modal_api_key": "sk-modal-..."
+}
+```
+
+Then select a `modal/` model in the model picker.
+
+**Details:**
+- Free promotional period until April 30, 2026
+- Model: GLM-5.1 FP8 (128k context, 16k max output)
+- No credit card required during the promotional period
+
 ### Mistral
 
 Add API key to `~/.pi/free.json` or environment variables:
@@ -233,6 +261,28 @@ Add API key to `~/.pi/free.json` or environment variables:
 export MISTRAL_API_KEY="..."
 ```
 
+### Qwen (1,000 free requests/day)
+
+Qwen provides free access to **Qwen Coder** (Qwen3.6-Plus) via OAuth device flow — no API key or credit card needed.
+
+```
+/login qwen
+```
+
+This command will:
+1. Open your browser to Qwen Studio's authorization page
+2. Display a device code (enter it if the browser doesn't pre-fill it)
+3. Wait for you to authorize in the browser
+4. Automatically complete login once approved
+
+Then select a `qwen/` model in the model picker.
+
+**Details:**
+- Free tier: 1,000 requests/day
+- Model: Qwen Coder (131k context, 16k max output)
+- No credit card required — just a free [Qwen Studio](https://chat.qwen.ai) account
+- To re-authenticate: `/logout qwen` then `/login qwen`
+
 ---
 
 ## Slash Commands
@@ -248,6 +298,8 @@ Each provider has toggle commands to switch between free and all models:
 | `/cline-toggle` | Toggle between free/all Cline models |
 | `/mistral-toggle` | Toggle between free/all Mistral models |
 | `/ollama-toggle` | Toggle between free/all Ollama models |
+| `/login qwen` | Authenticate with Qwen Studio (OAuth) |
+| `/logout qwen` | Clear Qwen credentials |
 
 **The toggle command:**
 - Switches between showing only free models vs. all available models
@@ -269,6 +321,7 @@ Create `~/.pi/free.json` in your home directory:
   "opencode_api_key": "YOUR_ZEN_KEY",
   "ollama_api_key": "YOUR_OLLAMA_KEY",
   "ollama_show_paid": true,
+  "modal_api_key": "YOUR_MODAL_KEY",
   "hidden_models": ["model-id-to-hide"]
 }
 ```
@@ -282,6 +335,31 @@ export NVIDIA_API_KEY="..."
 
 ---
 
+## Logging & Debugging
+
+pi-free now writes extension logs to:
+
+- **Windows:** `%USERPROFILE%\.pi\free.log`
+- **Linux/macOS:** `~/.pi/free.log`
+
+Useful env vars:
+
+```bash
+# Console log verbosity (default: error)
+LOG_LEVEL=debug
+
+# File log verbosity (default: debug)
+PI_FREE_LOG_LEVEL=debug
+
+# Custom log path (optional)
+PI_FREE_LOG_PATH=/tmp/pi-free.log
+
+# Disable file logging
+PI_FREE_FILE_LOG=false
+```
+
+---
+
 ## License
 
 MIT — See [LICENSE](LICENSE)

package/config.ts

@@ -6,10 +6,12 @@
  *   2. ~/.pi/free.json
  *
  * Per-provider paid model flags:
+ *   KILO_SHOW_PAID=true or kilo_show_paid: true
  *   OPENROUTER_SHOW_PAID=true or openrouter_show_paid: true
  *   NVIDIA_SHOW_PAID=true or nvidia_show_paid: true
  *   FIREWORKS_SHOW_PAID=true or fireworks_show_paid: true
  *   CLINE_SHOW_PAID=true or cline_show_paid: true
+ *   GO_SHOW_PAID=true or go_show_paid: true
  *   OLLAMA_SHOW_PAID=true or ollama_show_paid: true
  *
  * PI_FREE_KILO_FREE_ONLY=true — restrict Kilo to free models even after login.
@@ -25,17 +27,21 @@ interface PiFreeConfig {
 	openrouter_api_key?: string;
 	nvidia_api_key?: string;
 	opencode_api_key?: string;
+	opencode_go_api_key?: string;
 	fireworks_api_key?: string;
 	mistral_api_key?: string;
 	ollama_api_key?: string;
+	modal_api_key?: string;
 	kilo_free_only?: boolean;
 	hidden_models?: string[];
 	// Per-provider paid model flags
+	kilo_show_paid?: boolean;
 	openrouter_show_paid?: boolean;
 	nvidia_show_paid?: boolean;
 	fireworks_show_paid?: boolean;
 	cline_show_paid?: boolean;
 	zen_show_paid?: boolean;
+	go_show_paid?: boolean;
 	mistral_show_paid?: boolean;
 	ollama_show_paid?: boolean;
 }
@@ -44,16 +50,20 @@ const CONFIG_TEMPLATE: PiFreeConfig = {
 	openrouter_api_key: "",
 	nvidia_api_key: "",
 	opencode_api_key: "",
+	opencode_go_api_key: "",
 	fireworks_api_key: "",
 	mistral_api_key: "",
 	ollama_api_key: "",
+	modal_api_key: "",
 	kilo_free_only: false,
 	hidden_models: [],
+	kilo_show_paid: false,
 	openrouter_show_paid: false,
 	nvidia_show_paid: false,
 	fireworks_show_paid: false,
 	cline_show_paid: false,
 	zen_show_paid: false,
+	go_show_paid: false,
 	mistral_show_paid: false,
 	ollama_show_paid: false,
 };
@@ -123,6 +133,11 @@ function resolveBool(envKey: string, fileVal?: boolean): boolean {
 export const SHOW_PAID = process.env.PI_FREE_SHOW_PAID === "true";
 
 // Per-provider paid model flags - default to false (free-only) if not set
+export const KILO_SHOW_PAID = resolveBool(
+	"KILO_SHOW_PAID",
+	file.kilo_show_paid,
+);
+
 export const OPENROUTER_SHOW_PAID = resolveBool(
 	"OPENROUTER_SHOW_PAID",
 	file.openrouter_show_paid,
@@ -145,6 +160,8 @@ export const CLINE_SHOW_PAID = resolveBool(
 
 export const ZEN_SHOW_PAID = resolveBool("ZEN_SHOW_PAID", file.zen_show_paid);
 
+export const GO_SHOW_PAID = resolveBool("GO_SHOW_PAID", file.go_show_paid);
+
 export const MISTRAL_SHOW_PAID = resolveBool(
 	"MISTRAL_SHOW_PAID",
 	file.mistral_show_paid,
@@ -177,23 +194,31 @@ export const OPENCODE_API_KEY = resolve(
 	"OPENCODE_API_KEY",
 	file.opencode_api_key,
 );
+export const OPENCODE_GO_API_KEY = resolve(
+	"OPENCODE_GO_API_KEY",
+	file.opencode_go_api_key,
+);
 export const FIREWORKS_API_KEY = resolve(
 	"FIREWORKS_API_KEY",
 	file.fireworks_api_key,
 );
 export const MISTRAL_API_KEY = resolve("MISTRAL_API_KEY", file.mistral_api_key);
 export const OLLAMA_API_KEY = resolve("OLLAMA_API_KEY", file.ollama_api_key);
+export const MODAL_API_KEY = resolve("MODAL_API_KEY", file.modal_api_key);
 
 // Re-export provider names for consistency
 export {
 	PROVIDER_CLINE,
 	PROVIDER_FIREWORKS,
+	PROVIDER_GO,
 	PROVIDER_KILO,
 	PROVIDER_MISTRAL,
 	PROVIDER_NVIDIA,
 	PROVIDER_OLLAMA,
 	PROVIDER_OPENROUTER,
+	PROVIDER_QWEN,
 	PROVIDER_ZEN,
+	PROVIDER_MODAL,
 } from "./constants.ts";
 
 // =============================================================================

package/constants.ts

@@ -9,20 +9,28 @@
 
 export const PROVIDER_KILO = "kilo";
 export const PROVIDER_ZEN = "zen";
+export const PROVIDER_GO = "go";
 export const PROVIDER_OPENROUTER = "openrouter";
 export const PROVIDER_NVIDIA = "nvidia";
 export const PROVIDER_CLINE = "cline";
 export const PROVIDER_FIREWORKS = "fireworks";
 export const PROVIDER_OLLAMA = "ollama";
+export const PROVIDER_MISTRAL = "mistral";
+export const PROVIDER_QWEN = "qwen";
+export const PROVIDER_MODAL = "modal";
 
 export const ALL_PROVIDERS = [
 	PROVIDER_KILO,
 	PROVIDER_ZEN,
+	PROVIDER_GO,
 	PROVIDER_OPENROUTER,
 	PROVIDER_NVIDIA,
 	PROVIDER_CLINE,
 	PROVIDER_FIREWORKS,
+	PROVIDER_MISTRAL,
 	PROVIDER_OLLAMA,
+	PROVIDER_QWEN,
+	PROVIDER_MODAL,
 ] as const;
 
 // =============================================================================
@@ -31,11 +39,13 @@ export const ALL_PROVIDERS = [
 
 export const BASE_URL_KILO = "https://api.kilo.ai/api/gateway";
 export const BASE_URL_ZEN = "https://opencode.ai/zen/v1";
+export const BASE_URL_GO = "https://opencode.ai/zen/go/v1";
 export const BASE_URL_OPENROUTER = "https://openrouter.ai/api/v1";
 export const BASE_URL_NVIDIA = "https://integrate.api.nvidia.com/v1";
 export const BASE_URL_CLINE = "https://api.cline.bot/api/v1";
 export const BASE_URL_FIREWORKS = "https://api.fireworks.ai/inference/v1";
 export const BASE_URL_OLLAMA = "https://ollama.com/v1";
+export const BASE_URL_MODAL = "https://api.us-west-2.modal.direct/v1";
 
 // =============================================================================
 // External URLs
@@ -44,7 +54,11 @@ export const BASE_URL_OLLAMA = "https://ollama.com/v1";
 export const URL_MODELS_DEV = "https://models.dev/api.json";
 export const URL_KILO_TOS = "https://kilo.ai/terms";
 export const URL_ZEN_TOS = "https://opencode.ai/terms";
+export const URL_GO_TOS = "https://opencode.ai/terms";
 export const URL_CLINE_TOS = "https://cline.bot/tos";
+export const URL_QWEN_TOS = "https://terms.alicloud.com/";
+export const URL_MODAL_TOS = "https://modal.com/terms";
+export const BASE_URL_QWEN = "https://dashscope.aliyuncs.com/compatible-mode/v1";
 
 // =============================================================================
 // Cline auth
@@ -98,7 +112,6 @@ export const KILO_TOKEN_EXPIRATION_MS = 365 * 24 * 60 * 60 * 1000; // 1 year
 export const PROVIDER_GROQ = "groq";
 export const PROVIDER_TOGETHER = "together";
 export const PROVIDER_DEEPINFRA = "deepinfra";
-export const PROVIDER_MISTRAL = "mistral";
 export const PROVIDER_PERPLEXITY = "perplexity";
 export const PROVIDER_XAI = "xai";
 

package/lib/logger.ts

@@ -1,8 +1,16 @@
 /**
  * Structured logging utility.
  * Replaces console.log statements with namespaced, level-based logging.
+ *
+ * File logging:
+ * - Default file: ~/.pi/free.log
+ * - Override path: PI_FREE_LOG_PATH=/custom/path/free.log
+ * - Disable file logging: PI_FREE_FILE_LOG=false
  */
 
+import { appendFileSync, existsSync, mkdirSync } from "node:fs";
+import { dirname, join } from "node:path";
+
 export type LogLevel = "debug" | "info" | "warn" | "error";
 
 interface LogEntry {
@@ -20,25 +28,55 @@ const LOG_LEVELS: Record<LogLevel, number> = {
 	error: 3,
 };
 
-// Default to error-only. Set LOG_LEVEL=debug or LOG_LEVEL=info to see more.
+// Console default: error-only. Set LOG_LEVEL=debug or LOG_LEVEL=info to see more.
 const currentLevel: LogLevel = (process.env.LOG_LEVEL as LogLevel) || "error";
+// File default: debug (so we can inspect full behavior in ~/.pi/free.log).
+const fileLevel: LogLevel =
+	(process.env.PI_FREE_LOG_LEVEL as LogLevel) || "debug";
 
-function shouldLog(level: LogLevel): boolean {
-	return LOG_LEVELS[level] >= LOG_LEVELS[currentLevel];
+function shouldLog(level: LogLevel, minLevel: LogLevel): boolean {
+	return LOG_LEVELS[level] >= LOG_LEVELS[minLevel];
 }
 
 function formatMessage(entry: LogEntry): string {
-	const data = entry.data ? ` ${JSON.stringify(entry.data)}` : "";
+	let data = "";
+	if (entry.data) {
+		try {
+			data = ` ${JSON.stringify(entry.data)}`;
+		} catch {
+			data = " [unserializable-data]";
+		}
+	}
 	return `[${entry.timestamp}] [${entry.level.toUpperCase()}] [${entry.namespace}] ${entry.message}${data}`;
 }
 
+const HOME_DIR = process.env.HOME || process.env.USERPROFILE || "";
+const DEFAULT_LOG_PATH = join(HOME_DIR, ".pi", "free.log");
+const LOG_PATH = process.env.PI_FREE_LOG_PATH || DEFAULT_LOG_PATH;
+const FILE_LOG_ENABLED = process.env.PI_FREE_FILE_LOG !== "false";
+
+function appendToFile(line: string): void {
+	if (!FILE_LOG_ENABLED) return;
+	try {
+		const dir = dirname(LOG_PATH);
+		if (!existsSync(dir)) {
+			mkdirSync(dir, { recursive: true });
+		}
+		appendFileSync(LOG_PATH, `${line}\n`, "utf8");
+	} catch {
+		// Never throw from logger
+	}
+}
+
 function log(
 	level: LogLevel,
 	namespace: string,
 	message: string,
 	data?: Record<string, unknown>,
 ): void {
-	if (!shouldLog(level)) return;
+	const logToConsole = shouldLog(level, currentLevel);
+	const logToFile = shouldLog(level, fileLevel);
+	if (!logToConsole && !logToFile) return;
 
 	const entry: LogEntry = {
 		timestamp: new Date().toISOString(),
@@ -49,6 +87,13 @@ function log(
 	};
 
 	const formatted = formatMessage(entry);
+	if (logToFile) {
+		appendToFile(formatted);
+	}
+
+	if (!logToConsole) {
+		return;
+	}
 
 	switch (level) {
 		case "debug":