npm - free-coding-models - Versions diffs - 0.3.48 → 0.3.50 - Mend

free-coding-models 0.3.48 → 0.3.50

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.

Files changed (8) hide show

package/CHANGELOG.md +39 -8
package/README.md +32 -32
package/package.json +1 -1
package/sources.js +65 -63
package/src/command-palette.js +1 -1
package/src/key-handler.js +137 -20
package/web/dist/assets/{index-Dg9WC-oF.js → index-DOtGtGLl.js} +1 -1
package/web/dist/index.html +1 -1

package/CHANGELOG.md CHANGED Viewed

@@ -1,10 +1,41 @@
-## [0.3.48] - 2026-04-11
+## [0.3.50] - 2026-04-11
-### Fixed
-- **jcode model validation freeze (for real this time)** — jcode has a hardcoded model whitelist that rejects bare model names like `gpt-oss-120b` via `--model` flag. The v0.3.47 fix using `JCODE_MODEL` env var didn't actually work — jcode silently ignores it for the `openai-compatible` provider. The real fix uses two strategies:
-  1. **Native providers**: For providers that jcode supports natively (Groq, Cerebras, DeepInfra, Scaleway, Together, Hugging Face, Fireworks, Chutes, OpenRouter, Perplexity, ZAI, Mistral), we now use `--provider groq` instead of `--provider openai-compatible`. Native providers accept namespaced model names like `openai/gpt-oss-120b` without validation issues.
-  2. **OpenAI-compatible fallback**: For providers without a native jcode match (NVIDIA NIM, SambaNova, etc.), we keep `--provider openai-compatible` but ensure model IDs always have a namespace prefix (`openai/gpt-oss-120b` instead of `gpt-oss-120b`) and set a placeholder `OPENROUTER_API_KEY` to satisfy jcode's false credential check on namespaced models.
+### Changed
-### Added
-- **`JCODE_NATIVE_PROVIDERS` mapping** — New provider mapping table that routes 12 of our providers to jcode's native provider system, with correct env var names extracted from the jcode binary. This gives better compatibility and avoids the `openai-compatible` provider's model validation bugs.
-- **`ensureJcodeModelPrefix()` helper** — Ensures model IDs always have a namespace prefix (e.g. `gpt-oss-120b` → `openai/gpt-oss-120b`) so jcode's whitelist validation is bypassed.
+- **Providers reordered by generosity of free tier** — All 25 providers are now sorted from most generous to least generous in the README, TUI Settings page, and `D` key filter cycling. No more hunting for the best free option.
+- **Free tier limits corrected across all providers** — Verified and corrected free tier limits for every provider using live web research. Key corrections:
+  - **Groq**: 30 RPM, 1K-14.4K req/day (previously listed as "30-50 RPM per model")
+  - **Google AI Studio**: 15-60 RPM, 250-1.5K req/day (previously listed as "14.4K req/day, 30/min")
+  - **Together AI**: ❌ **No free tier** — requires $5 minimum purchase. Removed from the "free" recommendation.
+  - **iFlow**: ⚠️ **Shutting down April 17, 2026** — marked with warning in README and sources.js
+- **README subtitle updated** — Now says "ranked by generosity of free tier (most generous first)" instead of "ranked by SWE-bench"
+### Provider generosity ranking (most generous first)
+1. Groq (30 RPM, 1K-14.4K req/day)
+2. Cerebras (1M tokens/day)
+3. Google AI Studio (15-60 RPM, 250-1.5K req/day)
+4. NVIDIA NIM (~40 RPM)
+5. Cloudflare Workers AI (10K neurons/day)
+6. OpenRouter (50 req/day free)
+7. DeepInfra (200 concurrent requests)
+8. HuggingFace (~$0.10/month)
+9. Perplexity (~50 RPM tiered)
+10. SambaNova (generous dev quota)
+11. Fireworks AI ($1 credits)
+12. Hyperbolic ($1 credits)
+13. OVHcloud AI (2 req/min/IP free)
+14. Replicate (6 req/min free)
+15. Codestral (30 RPM, 2K req/day)
+16. ZAI (generous free quota)
+17. Scaleway (1M tokens)
+18. Alibaba DashScope (1M tokens/90 days)
+19. SiliconFlow (100 req/day + $1 credits)
+20. Rovo Dev CLI (5M tokens/day)
+21. Gemini CLI (1K req/day)
+22. Chutes AI (free community GPU)
+23. OpenCode Zen (free with account)
+24. Together AI (❌ no free tier)
+25. iFlow (⚠️ shutting down April 17, 2026)

package/README.md CHANGED Viewed

@@ -34,7 +34,7 @@ create a free account on one of the [providers](#-list-of-free-ai-providers)
   <a href="#-why-this-tool">💡 Why</a> •
   <a href="#-quick-start">⚡ Quick Start</a> •
   <a href="#-list-of-free-ai-providers">🟢 Providers</a> •
-  <a href="#-bonus-free-stuff">🎁 Bonus Free Stuff</a> •
+  <a href="#-other-free-ai-resources">🆓 Other Free AI Resources</a> •
   <a href="#-usage">🚀 Usage</a> •
   <a href="#-tui-keys">⌨️ TUI Keys</a> •
   <a href="#-features">✨ Features</a> •
@@ -72,43 +72,43 @@ It then writes the model you pick directly into your coding tool's config — so
 Create a free account on one provider below to get started:
-**238 coding models** across 25 providers, ranked by [SWE-bench Verified](https://www.swebench.com).
-| Provider | Models | Tier range | Free tier | Env var |
-|----------|--------|-----------|-----------|--------|
-| [NVIDIA NIM](https://build.nvidia.com) | 46 | S+ → C | 40 req/min (no credit card needed) | `NVIDIA_API_KEY` |
-| [OpenRouter](https://openrouter.ai/keys) | 25 | S+ → C | Free on :free: 50/day <$10, 1000/day ≥$10 (20 req/min) | `OPENROUTER_API_KEY` |
-| [Cloudflare Workers AI](https://dash.cloudflare.com) | 15 | S → B | Free: 10k neurons/day, text-gen 300 RPM | `CLOUDFLARE_API_TOKEN` + `CLOUDFLARE_ACCOUNT_ID` |
-| [SambaNova](https://sambanova.ai/developers) | 13 | S+ → B | Dev tier generous quota | `SAMBANOVA_API_KEY` |
-| [Hyperbolic](https://app.hyperbolic.ai/settings) | 13 | S+ → A- | $1 free trial credits | `HYPERBOLIC_API_KEY` |
-| [Together AI](https://api.together.ai/settings/api-keys) | 19 | S+ → A- | Credits/promos vary by account (check console) | `TOGETHER_API_KEY` |
-| [Scaleway](https://console.scaleway.com/iam/api-keys) | 10 | S+ → B+ | 1M free tokens | `SCALEWAY_API_KEY` |
-| [iFlow](https://platform.iflow.cn) | 11 | S+ → A+ | Free for individuals (no req limits, 7-day key expiry) | `IFLOW_API_KEY` |
-| [Alibaba DashScope](https://modelstudio.console.alibabacloud.com) | 11 | S+ → A | 1M free tokens per model (Singapore region, 90 days) | `DASHSCOPE_API_KEY` |
-| [Groq](https://console.groq.com/keys) | 8 | S → B | 30‑50 RPM per model (varies by model) | `GROQ_API_KEY` |
-| [Rovo Dev CLI](https://www.atlassian.com/rovo) | 5 | S+ | 5M tokens/day (beta) | CLI tool 🦘 |
-| [ZAI](https://z.ai) | 7 | S+ → S | Free tier (generous quota) | `ZAI_API_KEY` |
-| [OpenCode Zen](https://opencode.ai/zen) | 7 | S+ → A+ | Free with OpenCode account | Zen models ✨ |
-| [Google AI Studio](https://aistudio.google.com/apikey) | 6 | B+ → C | 14.4K req/day, 30/min | `GOOGLE_API_KEY` |
-| [SiliconFlow](https://cloud.siliconflow.cn/account/ak) | 6 | S+ → A | Free models: usually 100 RPM, varies by model | `SILICONFLOW_API_KEY` |
-| [Cerebras](https://cloud.cerebras.ai) | 4 | S+ → B | Generous free tier (developer tier 10× higher limits) | `CEREBRAS_API_KEY` |
-| [Perplexity API](https://www.perplexity.ai/settings/api) | 4 | A+ → B | Tiered limits by spend (default ~50 RPM) | `PERPLEXITY_API_KEY` |
-| [OVHcloud AI Endpoints](https://endpoints.ai.cloud.ovh.net) | 8 | S → B | Free sandbox: 2 req/min/IP (no key). 400 RPM with key | `OVH_AI_ENDPOINTS_ACCESS_TOKEN` |
-| [Chutes AI](https://chutes.ai) | 4 | S → A | Free (community GPU-powered, no credit card) | `CHUTES_API_KEY` |
-| [DeepInfra](https://deepinfra.com/login) | 4 | A- → B+ | 200 concurrent requests (default) | `DEEPINFRA_API_KEY` |
-| [Fireworks AI](https://fireworks.ai) | 4 | S → B+ | $1 credits – 10 req/min without payment | `FIREWORKS_API_KEY` |
-| [Gemini CLI](https://github.com/google-gemini/gemini-cli) | 3 | S+ → A+ | 1,000 req/day | CLI tool ♊ |
-| [Hugging Face](https://huggingface.com/settings/tokens) | 2 | S → B | Free monthly credits (~$0.10) | `HUGGINGFACE_API_KEY` |
-| [Replicate](https://replicate.com/account/api-tokens) | 2 | A- → B | 6 req/min (no payment) – up to 3,000 RPM with payment | `REPLICATE_API_TOKEN` |
-| [Mistral Codestral](https://codestral.mistral.ai) | 1 | B+ | 30 req/min, 2000/day | `CODESTRAL_API_KEY` |
+**238 coding models** across 25 providers, ranked by generosity of free tier (most generous first).
+| # | Provider | Models | Tier range | Free tier | Env var |
+|---|----------|--------|-----------|-----------|--------|
+| 1 | [Groq](https://console.groq.com/keys) | 8 | S → B | 30 RPM, 1K‑14.4K req/day (no credit card) | `GROQ_API_KEY` |
+| 2 | [Cerebras](https://cloud.cerebras.ai) | 4 | S+ → B | 30 RPM, 1M tokens/day (no credit card) | `CEREBRAS_API_KEY` |
+| 3 | [Google AI Studio](https://aistudio.google.com/apikey) | 6 | B+ → C | 15‑60 RPM, 250‑1.5K req/day (no credit card) | `GOOGLE_API_KEY` |
+| 4 | [NVIDIA NIM](https://build.nvidia.com) | 46 | S+ → C | ~40 RPM (no credit card) | `NVIDIA_API_KEY` |
+| 5 | [Cloudflare Workers AI](https://dash.cloudflare.com) | 15 | S → B | 10K neurons/day, 300 RPM (no credit card) | `CLOUDFLARE_API_TOKEN` + `CLOUDFLARE_ACCOUNT_ID` |
+| 6 | [OpenRouter](https://openrouter.ai/keys) | 25 | S+ → C | 50 req/day free, 1K/day with $10 spend | `OPENROUTER_API_KEY` |
+| 7 | [DeepInfra](https://deepinfra.com/login) | 4 | A- → B+ | 200 concurrent requests (no credit card) | `DEEPINFRA_API_KEY` |
+| 8 | [Hugging Face](https://huggingface.com/settings/tokens) | 2 | S → B | ~$0.10/month free credits | `HUGGINGFACE_API_KEY` |
+| 9 | [Perplexity API](https://www.perplexity.ai/settings/api) | 4 | A+ → B | ~50 RPM (tiered by spend) | `PERPLEXITY_API_KEY` |
+| 10 | [SambaNova](https://sambanova.ai/developers) | 13 | S+ → B | Dev tier generous quota (no credit card) | `SAMBANOVA_API_KEY` |
+| 11 | [Fireworks AI](https://fireworks.ai) | 4 | S → B+ | $1 free credits, 10 RPM without payment | `FIREWORKS_API_KEY` |
+| 12 | [Hyperbolic](https://app.hyperbolic.ai/settings) | 13 | S+ → A- | $1 free credits (permanent) | `HYPERBOLIC_API_KEY` |
+| 13 | [OVHcloud AI Endpoints](https://endpoints.ai.cloud.ovh.net) | 8 | S → B | 2 req/min/IP free, 400 RPM with key | `OVH_AI_ENDPOINTS_ACCESS_TOKEN` |
+| 14 | [Replicate](https://replicate.com/account/api-tokens) | 2 | A- → B | 6 req/min free, 3K RPM with payment | `REPLICATE_API_TOKEN` |
+| 15 | [Codestral](https://codestral.mistral.ai) | 1 | B+ | 30 RPM, 2K req/day (no credit card) | `CODESTRAL_API_KEY` |
+| 16 | [ZAI](https://z.ai) | 7 | S+ → S | Generous free quota (concurrency limited) | `ZAI_API_KEY` |
+| 17 | [Scaleway](https://console.scaleway.com/iam/api-keys) | 10 | S+ → B+ | 1M free tokens (permanent) | `SCALEWAY_API_KEY` |
+| 18 | [Alibaba DashScope](https://modelstudio.console.alibabacloud.com) | 11 | S+ → A | 1M free tokens/model (90 days) | `DASHSCOPE_API_KEY` |
+| 19 | [SiliconFlow](https://cloud.siliconflow.cn/account/ak) | 6 | S+ → A | 100 req/day + $1 free credits | `SILICONFLOW_API_KEY` |
+| 20 | [Rovo Dev CLI](https://www.atlassian.com/rovo) | 5 | S+ | 5M tokens/day (beta) | CLI tool 🦘 |
+| 21 | [Gemini CLI](https://github.com/google-gemini/gemini-cli) | 3 | S+ → A+ | 1,000 req/day (no credit card) | CLI tool ♊ |
+| 22 | [Chutes AI](https://chutes.ai) | 4 | S → A | Free community GPU (no credit card) | `CHUTES_API_KEY` |
+| 23 | [OpenCode Zen](https://opencode.ai/zen) | 7 | S+ → A+ | Free with OpenCode account | Zen models ✨ |
+| 24 | [Together AI](https://api.together.ai/settings/api-keys) | 19 | S+ → A- | ❌ No free tier — requires $5 purchase | `TOGETHER_API_KEY` |
+| 25 | [iFlow ⚠️](https://platform.iflow.cn) | 11 | S+ → A+ | Shutting down April 17, 2026 | `IFLOW_API_KEY` |
 > 💡 One key is enough. Add more at any time with **`P`** inside the TUI.
 ---
-## 🎁 Bonus Free Stuff
+## 🆓 Other Free AI Resources
-**Everything free that isn't in the CLI** — IDE extensions, coding agents, GitHub lists, trial credits, and more.
+**Curated free resources outside the CLI** — IDE extensions, coding agents, GitHub lists, and trial credits.
 ### 📚 Awesome Lists (curated by the community)

package/package.json CHANGED Viewed

@@ -1,6 +1,6 @@
 {
   "name": "free-coding-models",
-  "version": "0.3.48",
+  "version": "0.3.50",
   "description": "Find the fastest coding LLM models in seconds — ping free models from multiple providers, pick the best one for OpenCode, Cursor, or any AI coding assistant.",
   "keywords": [
     "nvidia",

package/sources.js CHANGED Viewed

@@ -480,12 +480,9 @@ export const opencodeZen = [
 // 📖 All sources combined - used by the main script
 // 📖 Each source has: name (display), url (API endpoint), models (array of model tuples)
+// 📖 Providers ordered by generosity of free tier (most generous first)
+// 📖 See README for full tier-by-tier comparison
 export const sources = {
-  nvidia: {
-    name: 'NIM',
-    url: 'https://integrate.api.nvidia.com/v1/chat/completions',
-    models: nvidiaNim,
-  },
   groq: {
     name: 'Groq',
     url: 'https://api.groq.com/openai/v1/chat/completions',
@@ -496,92 +493,91 @@ export const sources = {
     url: 'https://api.cerebras.ai/v1/chat/completions',
     models: cerebras,
   },
-  sambanova: {
-    name: 'SambaNova',
-    url: 'https://api.sambanova.ai/v1/chat/completions',
-    models: sambanova,
+  googleai: {
+    name: 'Google AI Studio',
+    url: 'https://generativelanguage.googleapis.com/v1beta/openai/chat/completions',
+    models: googleai,
+  },
+  nvidia: {
+    name: 'NIM',
+    url: 'https://integrate.api.nvidia.com/v1/chat/completions',
+    models: nvidiaNim,
+  },
+  cloudflare: {
+    name: 'Cloudflare AI',
+    url: 'https://api.cloudflare.com/client/v4/accounts/{account_id}/ai/v1/chat/completions',
+    models: cloudflare,
   },
   openrouter: {
     name: 'OpenRouter',
     url: 'https://openrouter.ai/api/v1/chat/completions',
     models: openrouter,
   },
+  deepinfra: {
+    name: 'DeepInfra',
+    url: 'https://api.deepinfra.com/v1/openai/chat/completions',
+    models: deepinfra,
+  },
   huggingface: {
     name: 'Hugging Face',
     url: 'https://router.huggingface.co/v1/chat/completions',
     models: huggingface,
   },
-  replicate: {
-    name: 'Replicate',
-    url: 'https://api.replicate.com/v1/predictions',
-    models: replicate,
+  perplexity: {
+    name: 'Perplexity',
+    url: 'https://api.perplexity.ai/chat/completions',
+    models: perplexity,
   },
-  deepinfra: {
-    name: 'DeepInfra',
-    url: 'https://api.deepinfra.com/v1/openai/chat/completions',
-    models: deepinfra,
+  sambanova: {
+    name: 'SambaNova',
+    url: 'https://api.sambanova.ai/v1/chat/completions',
+    models: sambanova,
   },
   fireworks: {
     name: 'Fireworks',
     url: 'https://api.fireworks.ai/inference/v1/chat/completions',
     models: fireworks,
   },
-  codestral: {
-    name: 'Codestral',
-    url: 'https://api.mistral.ai/v1/chat/completions',
-    models: codestral,
-  },
   hyperbolic: {
     name: 'Hyperbolic',
     url: 'https://api.hyperbolic.xyz/v1/chat/completions',
     models: hyperbolic,
   },
-  scaleway: {
-    name: 'Scaleway',
-    url: 'https://api.scaleway.ai/v1/chat/completions',
-    models: scaleway,
+  ovhcloud: {
+    name: 'OVHcloud AI',
+    url: 'https://oai.endpoints.kepler.ai.cloud.ovh.net/v1/chat/completions',
+    models: ovhcloud,
   },
-  googleai: {
-    name: 'Google AI',
-    url: 'https://generativelanguage.googleapis.com/v1beta/openai/chat/completions',
-    models: googleai,
+  replicate: {
+    name: 'Replicate',
+    url: 'https://api.replicate.com/v1/predictions',
+    models: replicate,
+  },
+  codestral: {
+    name: 'Codestral',
+    url: 'https://api.mistral.ai/v1/chat/completions',
+    models: codestral,
   },
   zai: {
     name: 'ZAI',
     url: 'https://api.z.ai/api/coding/paas/v4/chat/completions',
     models: zai,
   },
-  siliconflow: {
-    name: 'SiliconFlow',
-    url: 'https://api.siliconflow.com/v1/chat/completions',
-    models: siliconflow,
-  },
-  together: {
-    name: 'Together AI',
-    url: 'https://api.together.xyz/v1/chat/completions',
-    models: together,
-  },
-  cloudflare: {
-    name: 'Cloudflare AI',
-    url: 'https://api.cloudflare.com/client/v4/accounts/{account_id}/ai/v1/chat/completions',
-    models: cloudflare,
-  },
-  perplexity: {
-    name: 'Perplexity',
-    url: 'https://api.perplexity.ai/chat/completions',
-    models: perplexity,
+  scaleway: {
+    name: 'Scaleway',
+    url: 'https://api.scaleway.ai/v1/chat/completions',
+    models: scaleway,
   },
   qwen: {
-    name: 'Alibaba Cloud (DashScope)',
+    name: 'Alibaba DashScope',
     url: 'https://dashscope-intl.aliyuncs.com/compatible-mode/v1/chat/completions',
     models: qwen,
   },
-  iflow: {
-    name: 'iFlow',
-    url: 'https://apis.iflow.cn/v1/chat/completions',
-    models: iflow,
+  siliconflow: {
+    name: 'SiliconFlow',
+    url: 'https://api.siliconflow.com/v1/chat/completions',
+    models: siliconflow,
   },
-  // 📖 CLI-only tools (no API endpoint - launched directly)
   rovo: {
     name: 'Rovo Dev CLI',
     url: null, // CLI tool - no API endpoint
@@ -600,21 +596,27 @@ export const sources = {
     binary: 'gemini',
     checkArgs: ['--version'],
   },
+  chutes: {
+    name: 'Chutes AI',
+    url: 'https://chutes.ai/v1/chat/completions',
+    models: chutes,
+  },
   'opencode-zen': {
     name: 'OpenCode Zen',
     url: 'https://opencode.ai/zen/v1/chat/completions',
     models: opencodeZen,
     zenOnly: true,
   },
-  chutes: {
-    name: 'Chutes AI',
-    url: 'https://chutes.ai/v1/chat/completions',
-    models: chutes,
+  together: {
+    name: 'Together AI',
+    url: 'https://api.together.xyz/v1/chat/completions',
+    models: together,
   },
-  ovhcloud: {
-    name: 'OVHcloud AI 🆕',
-    url: 'https://oai.endpoints.kepler.ai.cloud.ovh.net/v1/chat/completions',
-    models: ovhcloud,
+  iflow: {
+    name: 'iFlow ⚠️',
+    url: 'https://apis.iflow.cn/v1/chat/completions',
+    models: iflow,
+    shutdownDate: '2026-04-17', // 📖 Shutting down April 17, 2026
   },
 }

package/src/command-palette.js CHANGED Viewed

@@ -40,7 +40,7 @@ const TOOL_MODE_COMMANDS = TOOL_MODE_ORDER.map((toolMode) => {
   const meta = TOOL_METADATA[toolMode] || { label: toolMode, emoji: '🧰' }
   return {
     id: `action-set-tool-${toolMode}`,
-    label: `${meta.emoji} ${meta.label}`,
+    label: meta.label,
     toolMode,
     icon: meta.emoji,
     description: TOOL_MODE_DESCRIPTIONS[toolMode] || 'Set this as the active launch target.',

package/src/key-handler.js CHANGED Viewed

@@ -55,12 +55,83 @@ const PROVIDER_TEST_MODEL_OVERRIDES = {
 const SETTINGS_TEST_MAX_ATTEMPTS = 10
 const SETTINGS_TEST_RETRY_DELAY_MS = 4000
+// 📖 PROVIDER_AUTH_ENDPOINTS maps provider keys to their auth-check URL + method.
+// 📖 For most providers this is the /models endpoint (returns 200=valid, 401=invalid).
+// 📖 Providers without an auth-check endpoint use null (falls back to chat completion ping).
+// 📖 Special cases:
+// 📖   - replicate: uses /v1/predictions (not /models) but needs a different payload
+// 📖   - cloudflare: no auth endpoint — only has chat completions, always uses ping fallback
+const PROVIDER_AUTH_ENDPOINTS = {
+  nvidia:       { url: 'https://api.nvidia.com/v1/account',           method: 'GET' },
+  groq:         { url: 'https://api.groq.com/v1/models',             method: 'GET' },
+  cerebras:     { url: 'https://api.cerebras.ai/v1/models',          method: 'GET' },
+  sambanova:    { url: 'https://api.sambanova.ai/v1/models',         method: 'GET' },
+  openrouter:   { url: 'https://openrouter.ai/api/v1/models',        method: 'GET' },
+  huggingface:  { url: 'https://router.huggingface.co/v1/models',    method: 'GET' },
+  deepinfra:    { url: 'https://api.deepinfra.com/v1/models',        method: 'GET' },
+  fireworks:   { url: 'https://api.fireworks.ai/v1/models',         method: 'GET' },
+  hyperbolic:   { url: 'https://api.hyperbolic.xyz/v1/models',       method: 'GET' },
+  scaleway:     { url: 'https://api.scaleway.ai/v1/models',          method: 'GET' },
+  siliconflow:  { url: 'https://api.siliconflow.com/v1/models',     method: 'GET' },
+  together:     { url: 'https://api.together.xyz/v1/models',        method: 'GET' },
+  perplexity:   { url: 'https://api.perplexity.ai/v1/models',       method: 'GET' },
+  chutes:       { url: 'https://chutes.ai/v1/models',               method: 'GET' },
+  ovhcloud:     { url: 'https://oai.endpoints.kepler.ai.cloud.ovh.net/v1/models', method: 'GET' },
+  qwen:         { url: 'https://dashscope-intl.aliyuncs.com/compatible-mode/v1/models', method: 'GET' },
+  iflow:        { url: 'https://apis.iflow.cn/v1/models',            method: 'GET' },
+  replicate:    null, // 📖 Replicate has no /models endpoint; use chat completions ping
+  cloudflare:   null, // 📖 Workers AI has no auth-check endpoint; use ping only
+  zai:          null, // 📖 ZAI undocumented; use ping only
+  googleai:     null, // 📖 Google AI Studio has no OpenAI-compatible /models; use ping
+  'opencode-zen': null, // 📖 OpenCode Zen uses OpenCode auth only; use ping
+  rovo:         null, // 📖 CLI tool — no API key
+  gemini:       null, // 📖 CLI tool — no API key
+}
 // 📖 Sleep helper kept local to this module so the Settings key test flow can
 // 📖 back off between retries without leaking timer logic into the rest of the TUI.
 function sleep(ms) {
   return new Promise((resolve) => setTimeout(resolve, ms))
 }
+// 📖 testProviderKeyDirect: Fast auth-only check using /v1/account or /v1/models.
+// 📖 Fires 3 parallel probes to get a fast decisive result (auth error vs timeout vs 200).
+// 📖 Returns { code, ms } from the first non-timeout response, or the best available.
+async function testProviderKeyDirect(apiKey, providerKey) {
+  const authConfig = PROVIDER_AUTH_ENDPOINTS[providerKey]
+  if (!authConfig) return null
+  const { url, method } = authConfig
+  const headers = { Authorization: `Bearer ${apiKey}` }
+  if (providerKey === 'openrouter') {
+    headers['HTTP-Referer'] = 'https://github.com/vava-nessa/free-coding-models'
+    headers['X-Title'] = 'free-coding-models'
+  }
+  const parallel = 3
+  const promises = Array.from({ length: parallel }, async () => {
+    const ctrl = new AbortController()
+    const timer = setTimeout(() => ctrl.abort(), 8000)
+    const t0 = performance.now()
+    try {
+      const resp = await fetch(url, { method, headers, signal: ctrl.signal })
+      return { code: resp.status, ms: Math.round(performance.now() - t0) }
+    } catch (err) {
+      const isTimeout = err.name === 'AbortError'
+      return { code: isTimeout ? '000' : 'ERR', ms: isTimeout ? 'TIMEOUT' : Math.round(performance.now() - t0) }
+    } finally {
+      clearTimeout(timer)
+    }
+  })
+  const results = await Promise.all(promises)
+  const success = results.find(r => r.code === 200)
+  if (success) return success
+  const authFailure = results.find(r => r.code === 401 || r.code === 403)
+  if (authFailure) return authFailure
+  return results[0]
+}
 /**
  * 📖 buildProviderModelsUrl derives the matching `/models` endpoint for providers
  * 📖 that expose an OpenAI-compatible model list next to `/chat/completions`.
@@ -471,7 +542,10 @@ export function createKeyHandler(ctx) {
   }
   // ─── Settings key test helper ───────────────────────────────────────────────
-  // 📖 Fires a single ping to the selected provider to verify the API key works.
+  // 📖 Verifies an API key by first doing a fast parallel auth-only probe (3×8s)
+  // 📖 to /v1/account or /v1/models, then falling back to chat completion pings.
+  // 📖 Auth-only result is decisive (200=valid, 401/403=invalid); only timeouts or
+  // 📖 providers without auth endpoints fall through to the ping-based approach.
   async function testProviderKey(providerKey) {
     const src = sources[providerKey]
     if (!src) return
@@ -484,6 +558,24 @@ export function createKeyHandler(ctx) {
       return
     }
+    // 📖 Fast path: parallel auth-only probes (3×8s) to /v1/account or /v1/models.
+    // 📖 200 = key valid and accepted. 401/403 = key rejected. null = no auth endpoint.
+    const authResult = await testProviderKeyDirect(testKey, providerKey)
+    if (authResult) {
+      if (authResult.code === 200) {
+        state.settingsTestResults[providerKey] = 'ok'
+        state.settingsTestDetails[providerKey] = buildProviderTestDetail(providerLabel, 'ok', [], `Auth-only probe returned HTTP 200.`)
+        return
+      }
+      if (authResult.code === 401 || authResult.code === 403) {
+        state.settingsTestResults[providerKey] = 'auth_error'
+        state.settingsTestDetails[providerKey] = buildProviderTestDetail(providerLabel, 'auth_error', [], `Auth probe returned HTTP ${authResult.code}.`)
+        return
+      }
+      // 📖 Timeout or ERR — fall through to ping-based approach below.
+    }
+    // 📖 Slow path: ping-based verification (providers without auth endpoint or timeouts).
     state.settingsTestResults[providerKey] = 'pending'
     state.settingsTestDetails[providerKey] = `Testing ${providerLabel} across up to ${SETTINGS_TEST_MAX_ATTEMPTS} probes...`
     const discoveredModelIds = []
@@ -508,7 +600,6 @@ export function createKeyHandler(ctx) {
           discoveryNote = `Live model discovery returned HTTP ${modelsResp.status}; falling back to the repo catalog.`
         }
       } catch (err) {
-        // 📖 Discovery failure is non-fatal; we still have repo-defined fallbacks.
         discoveryNote = `Live model discovery failed (${err?.name || 'error'}); falling back to the repo catalog.`
       }
     }
@@ -519,35 +610,46 @@ export function createKeyHandler(ctx) {
       state.settingsTestDetails[providerKey] = buildProviderTestDetail(providerLabel, 'fail', [], discoveryNote || 'No candidate model was available for probing.')
       return
     }
+    // 📖 Parallel ping burst: fire up to 5 probes simultaneously to get fast feedback.
+    const PARALLEL_PROBES = 5
     const attempts = []
+    let settled = false
-    for (let attemptIndex = 0; attemptIndex < SETTINGS_TEST_MAX_ATTEMPTS; attemptIndex++) {
-      const testModel = candidateModels[attemptIndex % candidateModels.length]
-      const { code } = await ping(testKey, testModel, providerKey, src.url)
-      attempts.push({ attempt: attemptIndex + 1, model: testModel, code })
+    while (!settled) {
+      const batch = []
+      for (let i = 0; i < PARALLEL_PROBES && attempts.length + batch.length < SETTINGS_TEST_MAX_ATTEMPTS; i++) {
+        const testModel = candidateModels[(attempts.length + batch.length) % candidateModels.length]
+        batch.push(ping(testKey, testModel, providerKey, src.url).then(({ code }) => ({ attempt: attempts.length + batch.length + 1, model: testModel, code })))
+      }
+      const batchResults = await Promise.all(batch)
+      attempts.push(...batchResults)
-      if (code === '200') {
+      // 📖 Check outcome after each parallel batch.
+      const outcome = classifyProviderTestOutcome(attempts.map(({ code }) => code))
+      if (outcome === 'ok') {
         state.settingsTestResults[providerKey] = 'ok'
         state.settingsTestDetails[providerKey] = buildProviderTestDetail(providerLabel, 'ok', attempts, discoveryNote)
-        return
+        settled = true
+        continue
       }
-      const outcome = classifyProviderTestOutcome(attempts.map(({ code: attemptCode }) => attemptCode))
       if (outcome === 'auth_error') {
         state.settingsTestResults[providerKey] = 'auth_error'
         state.settingsTestDetails[providerKey] = buildProviderTestDetail(providerLabel, 'auth_error', attempts, discoveryNote)
-        return
+        settled = true
+        continue
       }
-      if (attemptIndex < SETTINGS_TEST_MAX_ATTEMPTS - 1) {
-        state.settingsTestDetails[providerKey] = `Testing ${providerLabel}... probe ${attemptIndex + 1}/${SETTINGS_TEST_MAX_ATTEMPTS} failed on ${testModel} (${code}). Retrying in ${SETTINGS_TEST_RETRY_DELAY_MS / 1000}s.`
-        await sleep(SETTINGS_TEST_RETRY_DELAY_MS)
+      if (attempts.length >= SETTINGS_TEST_MAX_ATTEMPTS) {
+        state.settingsTestResults[providerKey] = outcome
+        state.settingsTestDetails[providerKey] = buildProviderTestDetail(providerLabel, outcome, attempts, discoveryNote)
+        settled = true
+        continue
       }
-    }
-    const finalOutcome = classifyProviderTestOutcome(attempts.map(({ code }) => code))
-    state.settingsTestResults[providerKey] = finalOutcome
-    state.settingsTestDetails[providerKey] = buildProviderTestDetail(providerLabel, finalOutcome, attempts, discoveryNote)
+      // 📖 Show progress between batches, then pause before next round.
+      state.settingsTestDetails[providerKey] = `Testing ${providerLabel}... ${attempts.length}/${SETTINGS_TEST_MAX_ATTEMPTS} probes tried. Retrying in ${SETTINGS_TEST_RETRY_DELAY_MS / 1000}s.`
+      await sleep(SETTINGS_TEST_RETRY_DELAY_MS)
+    }
   }
   // 📖 Manual update checker from settings; keeps status visible in maintenance row.
@@ -747,6 +849,20 @@ export function createKeyHandler(ctx) {
     state.settingsAddKeyMode = false
     state.settingsEditBuffer = ''
     state.settingsScrollOffset = 0
+    // 📖 Auto-test all configured API keys in parallel on Settings open.
+    // 📖 Each provider with a saved key fires a parallel auth probe batch immediately.
+    // 📖 The T key re-triggers a focused test on the selected row without clearing others.
+    const providerKeys = Object.keys(sources)
+    for (const pk of providerKeys) {
+      const testKey = getApiKey(state.config, pk)
+      if (testKey) {
+        // 📖 Fire and forget — update state as probes resolve.
+        testProviderKey(pk)
+      } else {
+        state.settingsTestResults[pk] = 'missing_key'
+      }
+    }
   }
   function openRecommendOverlay() {
@@ -2293,7 +2409,8 @@ export function createKeyHandler(ctx) {
     }
     // 📖 Alt+W: toggle footer visibility (collapse to single hint when hidden)
-    if (key.name === 'w' && key.alt && !key.ctrl && !key.meta) {
+    // 📖 Note: readline doesn't set key.alt=true for ALT combos; detect via str starting with \x1b (ESC)
+    if (key.name === 'w' && str && str.startsWith('\x1b') && !key.ctrl && !key.meta) {
       state.footerHidden = !state.footerHidden
       if (!state.config.settings || typeof state.config.settings !== 'object') state.config.settings = {}
       state.config.settings.footerHidden = state.footerHidden