pi-free 1.0.8 → 2.0.0

package/CHANGELOG.md

@@ -5,7 +5,113 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
-## [Unreleased]
+## [2.0.0] - 2026-04-23
+
+### Breaking Changes
+- **Removed Fireworks provider** — Fireworks is now a built-in Pi provider (added in pi 0.68.1), so the extension's Fireworks provider has been removed to avoid conflicts:
+  - Deleted `providers/fireworks/fireworks.ts` and `tests/fireworks.test.ts`
+  - Removed all Fireworks configuration options from `config.ts` (`fireworks_api_key`, `fireworks_show_paid`)
+  - Users should now use Pi's built-in Fireworks support with `FIREWORKS_API_KEY`
+
+- **Renamed Ollama provider to `ollama-cloud`** — Changed provider ID from `"ollama"` to `"ollama-cloud"` to avoid collision with Pi's built-in local Ollama provider:
+  - This prevents provider ID conflicts when both are registered
+  - All log messages and documentation now reference "Ollama Cloud"
+
+### Removed
+- **Dropped `@sinclair/typebox` peer dependency** — Pi 0.69.0 migrated from `@sinclair/typebox` to `typebox` 1.x. The extension didn't directly import this package, so it was removed from `peerDependencies` to avoid potential conflicts.
+
+### Fixed
+- **Ollama Cloud API endpoint** — Fixed broken Ollama Cloud integration:
+  - Changed `BASE_URL_OLLAMA` from `https://ollama.com` to `https://ollama.com/v1` — the OpenAI-compatible API endpoint
+  - Fixed model fetching to use `/v1/models` instead of `/api/tags` — ensures model IDs work with chat completions endpoint
+  - Previously calls went to HTML homepage instead of API endpoints, causing 404 errors
+
+### Removed
+- **Removed paid model warning on selection** — Deleted the `model_select` event handler that showed:
+  - `⚠️ Paid model selected (${model.id}). Use "/free off" to enable paid models.`
+  - This warning was redundant since the global `/free` toggle and provider toggles already control model visibility
+
+- **Removed pointless `/modal-toggle` command** — Modal provider only has 1 free model (GLM-5.1 FP8), so there was nothing meaningful to toggle:
+  - Added `skipToggle` option to `ProviderDefinition` and `ProviderSetupConfig` interfaces
+  - Modal provider now sets `skipToggle: true` to prevent toggle command creation
+
+### Changed
+- **Marked Qwen provider as fully deprecated** — Updated messaging to clarify the provider is broken:
+  - Changed model name from `"Qwen Coder — Free 1k/day"` to `"Qwen Coder — DEPRECATED (free tier discontinued)"`
+  - Updated all JSDoc comments to clearly state auth is broken and free tier is no longer available
+  - Provider remains for backward compatibility but should not be used
+
+### Added
+- **Cloudflare Workers AI provider** — New provider for Cloudflare's serverless GPU platform:
+  - 50+ open-source models: Llama 4, Mistral Small 3.1, Qwen 2.5/3, DeepSeek R1, Gemma 4, Kimi K2.5/2.6, and more
+  - **10,000 Neurons/day FREE tier** (resets daily at 00:00 UTC)
+  - **$0.011 per 1,000 Neurons** beyond free allocation
+  - Only requires `CLOUDFLARE_API_TOKEN` — account ID auto-derived from token
+  - Toggle with `/cloudflare-toggle`
+  - Create token at https://dash.cloudflare.com/profile/api-tokens
+
+- **Unified dynamic built-in providers module** — New `providers/dynamic-built-in/` module that dynamically fetches models from Pi's built-in providers when users have API keys:
+  - **Mistral** (`MISTRAL_API_KEY`) — Fetches from `api.mistral.ai/v1/models`
+  - **Groq** (`GROQ_API_KEY`) — Fetches from `api.groq.com/openai/v1/models`
+  - **Cerebras** (`CEREBRAS_API_KEY`) — Fetches from `api.cerebras.ai/v1/models`
+  - **xAI** (`XAI_API_KEY`) — Fetches from `api.x.ai/v1/models`
+  - **Hugging Face** (`HF_TOKEN` — optional) — Fetches public + authenticated models
+  - **OpenRouter** — Moved from `index.ts` to unified module with dynamic fetch
+  - All integrate with global `/free` toggle and have per-provider toggle commands (`/mistral-toggle`, `/groq-toggle`, etc.)
+
+- **Global `/free` toggle system** — New centralized free/paid filtering across ALL providers:
+  - `/free on/off/status` — Toggle free-only view globally
+  - `/free-providers` — Show free/paid model counts by provider
+  - `FREE_ONLY` config option and `PI_FREE_ONLY` environment variable
+  - Providers register via `registerWithGlobalToggle()` for unified filtering
+
+### Fixed
+- **Toggle commands now actually filter models from UI** — Previously, toggle commands only showed notifications but didn't remove paid models from the model picker:
+  - **OpenRouter (`/openrouter-toggle`)**: Now uses `registerProvider`/`unregisterProvider` to actually filter models from the picker UI
+  - **NVIDIA (`/nvidia-toggle`)**: Added dynamic `showPaid` parameter to `fetchNvidiaModels()` so toggle properly switches between free and paid model sets
+  - **Fireworks**: Removed broken toggle command — all models are paid with no free tier, so there was nothing to toggle
+
+### Added
+- **OpenRouter per-provider free model toggle** — Added `/openrouter-toggle` command for the built-in OpenRouter provider:
+  - `/openrouter-toggle` — Switch between showing only free models vs all models (including paid)
+  - New config flag `openrouter_show_paid` in `~/.pi/free.json` (default: `false`)
+  - Environment variable: `OPENROUTER_SHOW_PAID=true` to show paid models by default
+  - This brings OpenRouter (a built-in pi provider) in line with extension providers that have per-provider toggles
+
+### Deprecated
+- **Qwen provider** — The 1,000 requests/day free tier is no longer available from Qwen/DashScope. The provider code remains for backward compatibility but is now deprecated:
+  - Added `@deprecated` JSDoc tags to all Qwen-related exports
+  - Added deprecation warning when Qwen provider loads
+  - Added warning when `QWEN_SHOW_PAID` config is used
+  - Consider migrating to other free providers: Kilo, Cline, NVIDIA, or Modal
+
+### Added
+- **Go provider** — OpenCode Go subscription gateway (⚠️ paid only — $5 first month, then $10/month, no free tier) with models: GLM-5, Kimi K2.5, MiMo-V2-Pro, MiMo-V2-Omni, MiniMax M2.7, MiniMax M2.5
+  - Set `OPENCODE_GO_API_KEY` or `opencode_go_api_key` in `~/.pi/free.json`
+  - Toggle with `/go-toggle`
+
+### Fixed
+- **All providers now show Coding Index scores in model selector** — Added `enhanceWithCI()` to factory-based providers (nvidia, fireworks, mistral, modal, ollama) and cline. Now all providers display CI scores in `/models` command (pi-models extension).
+
+- **All providers now show in `--list-models`** — Providers (zen, openrouter, go) that registered models only in `session_start` were missing from `pi --list-models` which runs before session starts. Added immediate registration for these providers:
+  - **zen**: Added model caching to `~/.pi/provider-cache.json` for immediate registration + dynamic refresh
+  - **openrouter**: Immediate model registration at extension load (like kilo/cline)
+  - **go**: Immediate registration with static model list (no API to fetch from)
+  - All 11 providers now visible in `--list-models`
+
+### Changed
+- Updated README with clear free vs paid provider distinction (9 free + 2 paid-only: Go, Fireworks)
+- Added Go and Fireworks provider documentation under new "💳 Paid-Only Providers" section
+- Added `opencode_go_api_key` to config file template
+- Updated package.json description and keywords to include all 11 providers
+
+### Added
+- **Provider model cache** (`lib/provider-cache.ts`) — New utility for caching provider model lists to `~/.pi/provider-cache.json`. Used by zen provider for faster startup and offline access after first successful fetch.
+
+## [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
 

package/README.md

@@ -6,13 +6,13 @@ 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 10 different providers.**
+**pi-free is a Pi extension that unlocks free AI models from 9 providers — and adds 2 paid providers for convenience.**
 
 When you install pi-free, it:
 
-1. **Registers 9 AI providers** with Pi's model picker — OpenCode Zen, Kilo, OpenRouter, NVIDIA NIM, Cline, Mistral, Ollama Cloud, Qwen, and Modal
+1. **Registers 10+ AI providers** with Pi's model picker — 9 unique free-tier providers plus dynamic providers for Pi's built-in services when API keys are configured
 
-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
+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. Paid-only providers are hidden until you explicitly enable them.
 
 3. **Provides a toggle command** — Run `/{provider}-toggle` (e.g., `/zen-toggle`, `/kilo-toggle`) to switch between free-only mode and showing all models including paid ones
 
@@ -37,30 +37,48 @@ pi install git:github.com/apmantza/pi-free
 Start Pi and press `Ctrl+L` to open the model picker.
 
 Free models are shown by default — look for the provider prefixes:
+
+**✅ Offers Free Models (no usage limits, no payment required):**
 - `zen/` — OpenCode Zen models (no setup required)
 - `kilo/` — Kilo models (free models available immediately, more after `/login kilo`)
 - `openrouter/` — OpenRouter models (free account required)
-- `nvidia/` — NVIDIA NIM models (free API key required)
 - `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)
+
+**🔄 Freemium (free tier with limits, then paid):**
+- `nvidia/` — NVIDIA NIM models (1,000 free requests/month, then credits)
+- `cloudflare/` — Cloudflare Workers AI (10K Neurons/day free tier, then $0.011/1K Neurons)
+- `modal/` — GLM-5.1 FP8 via Modal (free promotional period until April 30, 2026)
+- `ollama/` — Ollama Cloud models (usage-based free tier, resets every 5 hours + 7 days)
+
+**🔧 Dynamic API Providers (free models when API key configured):**
+- `mistral/` — Mistral models (free models via API when `MISTRAL_API_KEY` set)
+- `groq/` — Groq models (free models via API when `GROQ_API_KEY` set)
+- `cerebras/` — Cerebras models (free models via API when `CEREBRAS_API_KEY` set)
+
+**💳 Paid Only (no free tier):**
+- `go/` — OpenCode Go models (requires subscription — $5 first month, then $10/month)
+
+**Note:** Fireworks is now a [built-in Pi provider](https://github.com/badlogic/pi-mono/blob/main/packages/coding-agent/CHANGELOG.md#0681---2026-04-22) — no extension needed. Set `FIREWORKS_API_KEY` to use it directly.
 
 ### 3. Toggle between free and paid models
 
 Want to see paid models too? Run the toggle command for your provider:
 
 ```
-/zen-toggle      # Toggle Zen free/paid models
-/kilo-toggle     # Toggle Kilo free/paid models
-/openrouter-toggle  # Toggle OpenRouter free/paid models
-/nvidia-toggle   # Toggle NVIDIA zero-cost/credit-costing models
-/cline-toggle    # Toggle Cline free/paid models
-/mistral-toggle  # Toggle Mistral free/paid models
-/ollama-toggle   # Toggle Ollama models (requires SHOW_PAID=true)
+/zen-toggle         # Toggle Zen (✅ offers free models)
+/kilo-toggle        # Toggle Kilo (✅ offers free models)
+/openrouter-toggle  # Toggle OpenRouter (✅ offers free models)
+/cline-toggle      # Toggle Cline (✅ offers free models)
+/mistral-toggle     # Toggle Mistral (🔧 dynamic - needs API key)
+/groq-toggle        # Toggle Groq (🔧 dynamic - needs API key)
+/cerebras-toggle   # Toggle Cerebras (🔧 dynamic - needs API key)
 ```
 
+**Notes:**
+- **Toggle commands are mainly for ✅ Offers Free Models providers** — to switch between "free models only" vs "show paid models too"
+- **🔄 Freemium providers** (NVIDIA, Cloudflare, Ollama, Modal) show all models by default — you manage your usage limits via their dashboards
+- **💳 Paid-only providers** (Go) have no toggle since all models require payment
+
 You'll see a notification like: `zen: showing free models` or `zen: showing all models (including paid)`
 
 ### 4. Add API keys for more providers (optional)
@@ -77,13 +95,15 @@ Add your API keys to this file:
 {
   "openrouter_api_key": "sk-or-v1-...",
   "nvidia_api_key": "nvapi-...",
+  "cloudflare_api_token": "...",
   "ollama_api_key": "...",
-  "fireworks_api_key": "...",
   "mistral_api_key": "...",
   "modal_api_key": "sk-modal-..."
 }
 ```
 
+**Note:** Fireworks is now a [built-in Pi provider](https://github.com/badlogic/pi-mono/blob/main/packages/coding-agent/CHANGELOG.md#0681---2026-04-22) — no extension needed. Set `FIREWORKS_API_KEY` to use it directly.
+
 Or set environment variables instead (same names, uppercase: `OPENROUTER_API_KEY`, `NVIDIA_API_KEY`, etc.)
 
 See the [Providers That Need Authentication](#providers-that-need-authentication) section below for detailed setup instructions per provider.
@@ -95,10 +115,8 @@ 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 |
 
 ---
 
@@ -141,7 +159,11 @@ Free tier resets every 5 hours + 7 days.
 
 ## Providers That Need Authentication
 
-Some providers require free accounts or OAuth to access their free tiers:
+Some providers require free accounts or OAuth to access their free tiers. **Go is a paid-only provider — it has no free tier.**
+
+---
+
+### 🆓 Free Providers
 
 ### Kilo (free models, more after login)
 
@@ -214,7 +236,31 @@ Or in `~/.pi/free.json`:
 
 Toggle anytime with `/nvidia-toggle`
 
-### Cline
+### Cloudflare Workers AI (10K Neurons/day Free Tier)
+
+Cloudflare provides **50+ open-source AI models** with a generous free tier:
+- **10,000 Neurons per day FREE** (resets daily at 00:00 UTC)
+- **$0.011 per 1,000 Neurons** beyond the free allocation
+
+Get your API token at [dash.cloudflare.com/profile/api-tokens](https://dash.cloudflare.com/profile/api-tokens):
+1. Create a token with "Cloudflare AI" → "Read" permission
+2. Or use "My Account" → "Read" for broader access
+
+**Setup:**
+```bash
+export CLOUDFLARE_API_TOKEN="your_token_here"
+```
+
+The account ID is automatically derived from your token. Optionally, you can also set:
+```bash
+export CLOUDFLARE_ACCOUNT_ID="your_account_id"  # Optional
+```
+
+**Models available:** Llama 4, Mistral Small 3.1, DeepSeek R1, Gemma 4, Kimi K2.5/2.6, and more.
+
+Toggle with `/cloudflare-toggle`
+
+### Cline (free account)
 
 Cline models appear immediately in the model picker. To use them, authenticate with Cline's free account:
 
@@ -230,7 +276,7 @@ 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 (GLM-5.1 FP8 — free promotional period 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:
 
@@ -253,7 +299,7 @@ Then select a `modal/` model in the model picker.
 - Model: GLM-5.1 FP8 (128k context, 16k max output)
 - No credit card required during the promotional period
 
-### Mistral
+### Mistral (free API key)
 
 Add API key to `~/.pi/free.json` or environment variables:
 
@@ -261,27 +307,29 @@ 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.
+### 💳 Paid-Only Providers
 
-```
-/login qwen
-```
+> **⚠️ These providers have no free tier. All usage incurs costs.**
 
-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
+### OpenCode Go (subscription — $5 first month, then $10/month)
 
-Then select a `qwen/` model in the model picker.
+Go provides access to curated open coding models via a monthly subscription. There is no free tier.
 
-**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`
+Set `OPENCODE_GO_API_KEY` (or `opencode_go_api_key` in `~/.pi/free.json`) and `GO_SHOW_PAID=true` to enable.
+
+**Models available:**
+- GLM-5
+- Kimi K2.5
+- MiMo-V2-Pro
+- MiMo-V2-Omni
+- MiniMax M2.7
+- MiniMax M2.5
+
+**Pricing:** $5 first month, then $10/month. See [opencode.ai/docs/go](https://opencode.ai/docs/go).
+
+Toggle with `/go-toggle`.
 
 ---
 
@@ -294,18 +342,19 @@ Each provider has toggle commands to switch between free and all models:
 | `/zen-toggle` | Toggle between free/all Zen models |
 | `/kilo-toggle` | Toggle between free/all Kilo models |
 | `/openrouter-toggle` | Toggle between free/all OpenRouter models |
-| `/nvidia-toggle` | Toggle between free/all NVIDIA 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 |
+| `/cline-toggle` | Toggle between free/all Cline models (✅ offers free models) |
+| `/mistral-toggle` | Toggle between free/all Mistral models (🔧 dynamic) |
+| `/groq-toggle` | Toggle between free/all Groq models (🔧 dynamic) |
+| `/cerebras-toggle` | Toggle between free/all Cerebras models (🔧 dynamic) |
 
 **The toggle command:**
-- Switches between showing only free models vs. all available models
+- **For ✅ Offers Free Models providers**: Switches between showing only free models vs. all available models (including paid)
+- **For 🔧 Dynamic API providers**: Filters the model list when you have an API key configured
 - **Persists your preference** to `~/.pi/free.json` for next startup
 - Shows a notification: "zen: showing free models" or "zen: showing all models (including paid)"
 
+**Note:** 🔄 Freemium providers (NVIDIA, Cloudflare, Ollama Cloud, Modal) don't have toggle commands — they show all models and you manage usage via their dashboards. 💳 Paid-only providers (Go) also have no toggle since all models require payment.
+
 ---
 
 ## Configuration
@@ -316,9 +365,9 @@ Create `~/.pi/free.json` in your home directory:
 {
   "openrouter_api_key": "YOUR_OPENROUTER_KEY",
   "nvidia_api_key": "YOUR_NVIDIA_KEY",
-  "fireworks_api_key": "YOUR_FIREWORKS_KEY",
   "mistral_api_key": "YOUR_MISTRAL_KEY",
   "opencode_api_key": "YOUR_ZEN_KEY",
+  "opencode_go_api_key": "YOUR_GO_KEY",
   "ollama_api_key": "YOUR_OLLAMA_KEY",
   "ollama_show_paid": true,
   "modal_api_key": "YOUR_MODAL_KEY",