free-coding-models 0.5.25 → 0.5.26

package/README.md

@@ -6,9 +6,9 @@
 
 <p align="center">
   <strong>Find the fastest free coding model in seconds</strong><br>
-  Track ~153 models across 16 trusted free or free-limited AI providers in real time<br><br>
+  Track ~146 models across 15 trusted free or free-limited AI providers in real time<br><br>
   <strong>Install Free API endpoints to your favorite AI coding tools:</strong><br>
-  OpenCode CLI / Desktop / WebUI, OpenClaw, Crush, Goose, Aider, Kilo CLI, Qwen Code, OpenHands, Amp, Hermes, Continue, Cline, Xcode, Pi, Rovo, Gemini and more...<br><br>
+  OpenCode CLI / Desktop / WebUI, OpenClaw, Crush, Goose, Aider, Kilo CLI, Qwen Code, OpenHands, Amp, Hermes, Continue, Cline, Xcode, Pi and more...<br><br>
   <strong>Use Kimi K2, DeepSeek V3, GPT-OSS, Qwen3, MiniMax M3, GLM, Llama 4, Gemma 4, Devstral and more — for free</strong>
 </p>
 
@@ -16,8 +16,8 @@
   <img src="https://img.shields.io/npm/v/free-coding-models?color=3d6b00&label=npm&logo=npm" alt="npm version" width="200"><br>
   <img src="https://img.shields.io/node/v/free-coding-models?color=3d6b00&logo=node.js" alt="node version" width="200"><br>
   <img src="https://img.shields.io/npm/l/free-coding-models?color=3d6b00" alt="license" width="200"><br>
-  <img src="https://img.shields.io/badge/models-154+-3d6b00?logo=nvidia" alt="models count" width="200"><br>
-  <img src="https://img.shields.io/badge/providers-16-1a56db" alt="providers count" width="200">
+  <img src="https://img.shields.io/badge/models-146-3d6b00?logo=nvidia" alt="models count" width="200"><br>
+  <img src="https://img.shields.io/badge/providers-15-1a56db" alt="providers count" width="200">
 </p>
 
 ```bash
@@ -91,12 +91,11 @@ Create a free account on one provider below to get started:
 | 12 | [ZAI](https://z.ai) | 2 | S | Free Flash models only | `ZAI_API_KEY` |
 | 13 | [Scaleway](https://console.scaleway.com/iam/api-keys) | 10 | S+ → B | 1M free tokens | `SCALEWAY_API_KEY` |
 | 14 | [Alibaba DashScope](https://modelstudio.console.alibabacloud.com) | 11 | S+ → A+ | 1M free tokens/model, Singapore, 90 days | `DASHSCOPE_API_KEY` |
-| 15 | [Gemini CLI](https://github.com/google-gemini/gemini-cli) | 7 | S+ → A | 1,000 req/day with personal Google account | CLI tool ♊ |
-| 16 | [OpenCode Zen](https://opencode.ai/zen) | 6 | S+ → A | Free with OpenCode account | Zen models ✨ |
+| 15 | [OpenCode Zen](https://opencode.ai/zen) | 6 | S+ → A | Free with OpenCode account | Zen models ✨ |
 
 > 💡 One key is enough. Add more at any time with **`P`** inside the TUI.
 
-> 🧹 Audit cleanup: `iFlow` was removed because it shut down on April 17, 2026. `Together AI`, `Perplexity API`, `DeepInfra`, `Replicate`, `Fireworks`, `Hyperbolic`, `Hugging Face`, `SiliconFlow`, `Chutes AI`, and `Rovo` were removed from the active free catalog because they are paid, trial-credit only, too tiny to be useful, unclear as a stable free API, or tool-specific rather than a generally usable free provider.
+> 🧹 Audit cleanup: `iFlow` was removed because it shut down on April 17, 2026. `Together AI`, `Perplexity API`, `DeepInfra`, `Replicate`, `Fireworks`, `Hyperbolic`, `Hugging Face`, `SiliconFlow`, `Chutes AI` were removed from the active free catalog because they are paid, trial-credit only, too tiny to be useful, unclear as a stable free API, or tool-specific rather than a generally usable free provider. `Rovo` and `Gemini CLI` were also wiped out as tool integrations (CLI-only, not generally usable free providers).
 
 ---
 
@@ -425,8 +424,6 @@ Routing behavior:
 | `--cline` | 🧠 Cline |
 | `--xcode` | 🛠️ Xcode Intelligence |
 | `--pi` | π Pi |
-| `--rovo` | 🦘 Rovo Dev CLI |
-| `--gemini` | ♊ Gemini CLI |
 | `--copilot` | 🤖 Copilot CLI |
 | `--forgecode` | 🔥 ForgeCode |
 
@@ -434,22 +431,6 @@ Press **`Z`** in the TUI to cycle between tools without restarting.
 
 ### CLI-Only Tools
 
-**🦘 Rovo Dev CLI**
-- Provider: [Atlassian Rovo](https://www.atlassian.com/rovo)
-- Install: [Installation Guide](https://support.atlassian.com/rovo/docs/install-and-run-rovo-dev-cli-on-your-device/)
-- Free tier: conditional Atlassian/Rovo allowance, not a general free API provider
-- Catalog status: not included in the active free-provider table because access depends on Rovo availability for your account/site
-- Launch: `free-coding-models --rovo` or press `Z` until Rovo mode
-- Features: Jira/Confluence integration, MCP server support
-
-**♊ Gemini CLI**
-- Provider: [Google Gemini](https://github.com/google-gemini/gemini-cli)
-- Install: `npm install -g @google/gemini-cli`
-- Free tier: 1,000 requests/day (personal Google account, no credit card)
-- Models: Gemini 3.1 Pro Preview, Gemini 3 Flash Preview, Gemini 2.5 Pro, Gemini 2.5 Flash
-- Launch: `free-coding-models --gemini` or press `Z` until Gemini mode
-- Features: OpenAI-compatible API support, MCP server support, Google Search grounding
-
 **Note:** When launching these tools via `Z` key or command palette, if the current mode doesn't match the tool, you'll see a confirmation alert asking to switch to the correct tool before launching.
 
 ### OpenCode Zen Free Models
@@ -472,10 +453,8 @@ When a tool mode is active (via `Z`), models incompatible with that tool are hig
 
 | Model Type | Compatible Tools |
 |------------|-----------------|
-| Regular (NVIDIA, Groq, etc.) | All tools except 🦘 Rovo and ♊ Gemini |
-| Rovo | 🦘 Rovo Dev CLI only |
-| Gemini | ♊ Gemini CLI only |
-| OpenCode Zen | 📦 OpenCode CLI and 📦 OpenCode Desktop only |
+| Regular (NVIDIA, Groq, etc.) | All tools |
+| OpenCode Zen | All tools (OpenAI-compatible endpoint) |
 
 → **[Full flags reference](./docs/flags.md)**
 
@@ -534,7 +513,7 @@ When a tool mode is active (via `Z`), models incompatible with that tool are hig
 
 ## ✨ Features
 
-- **Parallel pings** — all ~165 API/Zen-callable models tested simultaneously via native `fetch` (~170 total cataloged models including CLI-only Gemini rows)
+- **Parallel pings** — all ~150 API/Zen-callable models tested simultaneously via native `fetch` (~152 total cataloged models)
 - **AI benchmark columns** — `Ctrl+A` benchmarks the selected model, `Ctrl+U` benchmarks visible models, and results split cleanly into **AI Latency** plus **TPS**. Settings includes an opt-in **Startup AI Speed Scan** toggle to run the global benchmark automatically after launch.
 - **Tiny verdict indicator** — the first `❔` column mirrors the full Verdict as a compact emoji (`🟩`, `🟢`, `🟡`, `🟠`, etc.) and sorts by the same verdict order.
 - **Adaptive monitoring** — 2s burst for 60s → 10s normal → 30s idle
@@ -724,8 +703,6 @@ Telemetry is enabled by default and can be disabled with any of the following:
 | IDE | Free tier | Credit card |
 |-----|-----------|-------------|
 | [Qwen Code](https://github.com/QwenLM/qwen-code) | 2,000 requests/day | No |
-| [Rovo Dev CLI](https://www.atlassian.com/rovo) | Conditional Atlassian/Rovo allowance | No |
-| [Gemini CLI](https://github.com/google-gemini/gemini-cli) | 1,000 requests/day | No |
 | [Jules](https://jules.google/) | 15 tasks/day | No |
 | [AWS Kiro](https://kiro.dev/) | 50 credits/month | No |
 | [Trae](https://trae.ai/) | 10 fast + 50 slow requests/month | No |

package/changelog/v0.5.26.md

@@ -0,0 +1,21 @@
+# Changelog v0.5.26 - 2026-06-11
+
+### Removed
+- **Gemini CLI** wiped out as a tool integration across the entire stack (TUI cycle `Z`, command palette, theme, key handler, provider metadata, tool bootstrap, tool launchers, benchmark, endpoint installer, audit state, docker-compose, docs)
+- **Rovo Dev CLI** wiped out the same way — block removed from `tool-metadata.js`, `tool-bootstrap.js`, `provider-metadata.js`, `tool-launchers.js` (incl. `writeRovoConfig`), `benchmark.js`, `endpoint-installer.js`, `utils.js`, `command-palette.js`, `theme.js`, `key-handler.js`, `app.js`, plus the `--rovo` CLI flag
+- Both removed from the website's social proof bar (`Gemini CLI` and `Rovo Dev` chips)
+- `cliOnly` tool concept collapsed — no remaining CLI-only providers, so `getCompatibleTools` simplified to return `REGULAR_TOOLS` for both `regular` and `opencode-zen`
+
+### Changed
+- `sources.js` JSDoc `@exports` no longer lists the stale `gemini` provider
+- `audit_state.json` clone entry `gemini` removed (was a fingerprint duplicate of `googleai`)
+- `.pi/skills/update_models/SKILL.md`: `googleai` audit diff now applies directly to `sources.js` (no more `gemini` clone)
+- `docker-compose.yml`: dead `GEMINI_API_KEY` env var removed (not referenced anywhere in code)
+- `README.md`: tool list, compatibility table, IDE table, model count all updated
+- `tui/overlays.js` `Z` cycle help: `🦘 Rovo → ♊ Gemini` segment dropped
+
+### Tests
+- `tool compatibility matrix` test suite updated: removed 5 rovo/gemini-specific assertions, kept Zen compatibility coverage
+- `sortResultsWithPinnedFavorites` mock data swapped from `rovo` fixtures to `caveman` fixtures (caveman is also cliOnly-equivalent for sort order testing)
+- `benchmarkModel` no longer has a `providerKey === 'rovo'` UNSUPPORTED guard test (guard itself was removed with the wipeout)
+- 520/520 tests pass after refactor

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "free-coding-models",
-  "version": "0.5.25",
+  "version": "0.5.26",
   "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

@@ -29,7 +29,7 @@
  *   📖 Secondary: https://swe-rebench.com (independent evals, scores are lower)
  *   📖 Leaderboard tracker: https://www.marc0.dev/en/leaderboard
  *
- *   @exports nvidiaNim, groq, cerebras, sambanova, openrouter, githubModels, mistral, codestral, scaleway, googleai, zai, qwen, cloudflare, ovhcloud, gemini, opencodeZen — model arrays per active provider
+ *   @exports nvidiaNim, groq, cerebras, sambanova, openrouter, githubModels, mistral, codestral, scaleway, googleai, zai, qwen, cloudflare, ovhcloud, opencodeZen — model arrays per active provider
  *   @exports sources — map of active free/free-limited providers, each with { name, url, models }
 
  *   @exports MODELS — flat array of [modelId, label, tier, sweScore, ctx, providerKey]
@@ -339,25 +339,7 @@ export const ovhcloud = [
   // Fix (2026-05-26): Qwen3.5-9B ctx 128k→262k, Mistral-Small ctx 131k→128k, Mistral-Nemo ctx 128k→118k, Mistral-7B ctx 32k→127k
 ]
 
-// 📖 Gemini CLI source - https://github.com/google-gemini/gemini-cli
-// 📖 CLI tool with OpenAI-compatible API support
-// 📖 Install: npm install -g @google/gemini-cli
-// 📖 Free tier: 1,000 req/day with personal Google account (no credit card)
-// 📖 Models track Google AI Studio IDs; no stale google/ prefix.
-// 📖 Supports custom OpenAI-compatible providers via GEMINI_API_BASE_URL
-export const gemini = [
-  ['gemini-3.5-flash',                   'Gemini 3.5 Flash',             'S+', '-',     '1M'],
-  ['gemini-3.1-pro-preview',            'Gemini 3.1 Pro Preview',       'S+', '78.0%', '1M'],
-  ['gemini-3-flash-preview',            'Gemini 3 Flash Preview',       'S',  '65.0%', '1M'],
-  ['gemini-3.1-flash-lite',             'Gemini 3.1 Flash Lite',        'A+', '55.0%', '1M'],
-  // ⚠️ DEPRECATED — shutdown Oct 16, 2026
-  ['gemini-2.5-pro',                    'Gemini 2.5 Pro',               'S+', '63.2%', '1M'],
-  // ⚠️ DEPRECATED — shutdown Oct 16, 2026
-  ['gemini-2.5-flash',                  'Gemini 2.5 Flash',             'A+', '50.0%', '1M'],
-  // ⚠️ DEPRECATED — shutdown Oct 16, 2026
-  ['gemini-2.5-flash-lite',             'Gemini 2.5 Flash Lite',        'A',  '42.0%', '1M'],
-  // Removed (2026-05-26): gemini-3.1-flash-lite-preview (endpoint shutdown May 25, 2026)
-]
+
 
 // 📖 OpenCode Zen free models — hosted AI gateway accessed through OpenCode CLI/Desktop
 // 📖 Endpoint: https://opencode.ai/zen/v1/... — requires OpenCode Zen API key
@@ -456,15 +438,7 @@ export const sources = {
     url: 'https://dashscope-intl.aliyuncs.com/compatible-mode/v1/chat/completions',
     models: qwen,
   },
-  gemini: {
-    name: 'Gemini CLI',
-    url: null, // CLI tool - no API endpoint (can use OpenAI-compatible via env)
-    models: gemini,
-    cliOnly: true,
-    installUrl: 'https://github.com/google-gemini/gemini-cli',
-    binary: 'gemini',
-    checkArgs: ['--version'],
-  },
+
   'opencode-zen': {
     name: 'OpenCode Zen',
     url: 'https://opencode.ai/zen/v1/chat/completions',

package/src/core/benchmark.js

@@ -257,16 +257,6 @@ async function benchmarkSingleAttempt({ apiKey, modelId, providerKey, url, timeo
 // 📖 Returns on failure (all attempts exhausted):
 // 📖   { ok: false, code, totalMs, error, retries }
 export async function benchmarkModel({ apiKey, modelId, providerKey, url, timeoutMs = BENCHMARK_TIMEOUT_MS, maxRetries = BENCHMARK_MAX_RETRIES, retryDelayMs = BENCHMARK_RETRY_DELAY_MS }) {
-  // 📖 Guard: unsupported providers that don't do chat completions
-  if (providerKey === 'rovo' || providerKey === 'gemini' || providerKey === 'opencode-zen') {
-    return {
-      ok: false,
-      code: 'UNSUPPORTED',
-      totalMs: 0,
-      error: 'Provider does not support chat completions',
-      retries: 0,
-    }
-  }
 
   let lastResult = null
 

package/src/core/config.js

@@ -132,7 +132,6 @@ const ENV_VARS = {
   cloudflare: ['CLOUDFLARE_API_TOKEN', 'CLOUDFLARE_API_KEY'],
   qwen:       'DASHSCOPE_API_KEY',
   zai:        'ZAI_API_KEY',
-  gemini:     'GEMINI_API_KEY',
   ovhcloud:   'OVH_AI_ENDPOINTS_ACCESS_TOKEN',
   'opencode-zen': 'OPENCODE_ZEN_API_KEY',
 }

package/src/core/endpoint-installer.js

@@ -48,10 +48,9 @@ import { getApiKey, saveConfig } from './config.js'
 import { ENV_VAR_NAMES, PROVIDER_METADATA } from './provider-metadata.js'
 import { getToolMeta } from './tool-metadata.js'
 
-// 📖 CLI-only providers (rovo, gemini) cannot be installed into other tools — they manage their own auth.
 // 📖 replicate uses /v1/predictions (not /chat/completions), so it's not OpenAI-compatible.
 // 📖 zai and opencode-zen ARE OpenAI-compatible and CAN be installed into any tool.
-const DIRECT_INSTALL_UNSUPPORTED_PROVIDERS = new Set(['replicate', 'rovo', 'gemini'])
+const DIRECT_INSTALL_UNSUPPORTED_PROVIDERS = new Set(['replicate'])
 // 📖 Install Endpoints only lists tools whose persisted config shape is actually supported here.
 // 📖 Launch-only tools stay out: the Web dashboard configures endpoints, it never starts CLIs.
 const INSTALL_TARGET_MODES = ['opencode', 'opencode-desktop', 'opencode-web', 'openclaw', 'crush', 'goose', 'pi', 'aider', 'qwen', 'openhands', 'amp', 'forgecode', 'fcm_router']

package/src/core/legacy-proxy-cleanup.js

@@ -27,9 +27,9 @@ import { existsSync, readFileSync, writeFileSync, unlinkSync, readdirSync } from
 import { homedir } from 'node:os'
 import { join } from 'node:path'
 
-const LEGACY_TOOL_MODES = new Set(['claude-code', 'codex', 'gemini'])
+const LEGACY_TOOL_MODES = new Set(['claude-code', 'codex'])
 const LEGACY_RUNTIME_FILES = ['daemon.json', 'daemon-stdout.log', 'daemon-stderr.log', 'request-log.jsonl']
-const LEGACY_ENV_FILES = ['.fcm-claude-code-env', '.fcm-codex-env', '.fcm-gemini-env']
+const LEGACY_ENV_FILES = ['.fcm-claude-code-env', '.fcm-codex-env']
 
 function getDefaultPaths(homeDir) {
   return {

package/src/core/provider-metadata.js

@@ -52,7 +52,6 @@ export const ENV_VAR_NAMES = {
   googleai:   'GOOGLE_API_KEY',
   cloudflare: 'CLOUDFLARE_API_TOKEN',
   zai:        'ZAI_API_KEY',
-  gemini:     'GEMINI_API_KEY',
   ovhcloud:   'OVH_AI_ENDPOINTS_ACCESS_TOKEN',
   qwen:       'DASHSCOPE_API_KEY',
   'opencode-zen': 'OPENCODE_ZEN_API_KEY',
@@ -236,22 +235,6 @@ export const PROVIDER_METADATA = {
     signupHint: 'Install @mariozechner/pi-coding-agent and set ANTHROPIC_API_KEY',
     rateLimits: 'Depends on provider subscription (e.g., Anthropic, OpenAI)',
   },
-  rovo: {
-    label: 'Rovo Dev CLI',
-    color: chalk.rgb(148, 163, 184), // slate blue
-    signupUrl: 'https://www.atlassian.com/rovo',
-    signupHint: 'Install ACLI and run: acli rovodev auth login',
-    rateLimits: 'Free tier: 5M tokens/day (beta, requires Atlassian account)',
-    cliOnly: true,
-  },
-  gemini: {
-    label: 'Gemini CLI',
-    color: chalk.rgb(66, 165, 245), // blue
-    signupUrl: 'https://github.com/google-gemini/gemini-cli',
-    signupHint: 'Install: npm install -g @google/gemini-cli',
-    rateLimits: 'Free tier: 1,000 req/day with personal Google account',
-    cliOnly: true,
-  },
   'opencode-zen': {
     label: 'OpenCode Zen',
     color: chalk.rgb(139, 92, 246), // violet — distinctive from other providers

package/src/core/quota-capabilities.js

@@ -53,7 +53,6 @@ export const PROVIDER_CAPABILITIES = {
   // Providers with no reliable quota signal
   cloudflare: { telemetryType: 'unknown', supportsEndpoint: false, usageDisplay: 'ok', resetCadence: 'daily' },
   zai: { telemetryType: 'unknown', supportsEndpoint: false, usageDisplay: 'ok', resetCadence: 'none' },
-  gemini: { telemetryType: 'unknown', supportsEndpoint: false, usageDisplay: 'ok', resetCadence: 'daily' },
   'opencode-zen': { telemetryType: 'unknown', supportsEndpoint: false, usageDisplay: 'ok', resetCadence: 'unknown' },
 }
 

package/src/core/tool-bootstrap.js

@@ -227,17 +227,6 @@ export const TOOL_BOOTSTRAP_METADATA = {
       },
     },
   },
-  rovo: {
-    binary: 'acli',
-    docsUrl: 'https://support.atlassian.com/rovo/docs/install-and-run-rovo-dev-cli-on-your-device/',
-    install: {
-      default: {
-        shellCommand: 'npm install -g acli',
-        summary: 'Rovo Dev CLI requires ACLI installation. Visit the documentation for platform-specific instructions.',
-        note: 'Rovo is an Atlassian tool that requires an Atlassian account with Rovo Dev activated.',
-      },
-    },
-  },
   'continue': {
     binary: 'cn',
     docsUrl: 'https://docs.continue.dev/cli/overview',
@@ -279,17 +268,6 @@ export const TOOL_BOOTSTRAP_METADATA = {
       win32: 'Hermes Agent does not support native Windows. Use WSL2 instead.',
     },
   },
-  gemini: {
-    binary: 'gemini',
-    docsUrl: 'https://github.com/google-gemini/gemini-cli',
-    install: {
-      default: {
-        shellCommand: 'npm install -g @google/gemini-cli',
-        summary: 'Install Gemini CLI globally via npm.',
-        note: 'After installation, run `gemini` to authenticate with your Google account.',
-      },
-    },
-  },
   caveman: {
     binary: 'caveman',
     docsUrl: 'https://github.com/JuliusBrussee/caveman-code',

package/src/core/tool-launchers.js

@@ -427,29 +427,6 @@ function writeOpenHandsEnv(model, apiKey, baseUrl, paths = getDefaultToolPaths()
   return { filePath, backupPath }
 }
 
-/**
- * 📖 writeRovoConfig - Configure Rovo Dev CLI model selection
- *
- * Rovo Dev CLI uses ~/.rovodev/config.yml for configuration.
- * We write the model ID to the config file before launching.
- *
- * @param {Object} model - Selected model with modelId
- * @param {string} configPath - Path to Rovo config file
- * @returns {{ filePath: string, backupPath: string | null }}
- */
-function writeRovoConfig(model, configPath = join(homedir(), '.rovodev', 'config.yml')) {
-  const backupPath = backupIfExists(configPath)
-  const config = {
-    agent: {
-      modelId: model.modelId,
-    },
-  }
-
-  ensureDir(configPath)
-  writeFileSync(configPath, `agent:\n  modelId: "${model.modelId}"\n`)
-  return { filePath: configPath, backupPath }
-}
-
 // 📖 writeContinueConfig — write ~/.continue/config.yaml with the selected model.
 // 📖 Continue CLI uses YAML config with `provider: openai` for OpenAI-compatible endpoints.
 function writeContinueConfig(model, apiKey, baseUrl, paths = getDefaultToolPaths()) {
@@ -606,33 +583,6 @@ function restartHermesGateway() {
   spawnSync(hermesBin, ['gateway', 'restart'], { stdio: 'ignore', timeout: 10000 })
 }
 
-/**
- * 📖 buildGeminiEnv - Build environment variables for Gemini CLI
- *
- * Gemini CLI supports OpenAI-compatible APIs via environment variables:
- * - GEMINI_API_BASE_URL: Custom API endpoint
- * - GEMINI_API_KEY: API key for custom endpoint
- *
- * @param {Object} model - Selected model with providerKey
- * @param {Object} config - Full app config
- * @param {Object} options - Env options
- * @returns {NodeJS.ProcessEnv}
- */
-function buildGeminiEnv(model, config, options = {}) {
-  const providerKey = model.providerKey || 'gemini'
-  const apiKey = getApiKey(config, providerKey)
-  const baseUrl = getProviderBaseUrl(providerKey)
-
-  const env = cloneInheritedEnv(process.env, SANITIZED_TOOL_ENV_KEYS)
-
-  // If we have a custom API key and base URL, configure OpenAI-compatible mode
-  if (apiKey && baseUrl && options.includeProviderEnv) {
-    env.GEMINI_API_BASE_URL = baseUrl
-    env.GEMINI_API_KEY = apiKey
-  }
-
-  return env
-}
 
 /**
  * 📖 buildCavemanEnv - Build environment variables for Caveman Code
@@ -643,7 +593,7 @@ function buildGeminiEnv(model, config, options = {}) {
  *
  * Supported env vars (from caveman-code source):
  * - ANTHROPIC_API_KEY, OPENAI_API_KEY, GROQ_API_KEY, CEREBRAS_API_KEY
- * - MISTRAL_API_KEY, GOOGLE_API_KEY, GEMINI_API_KEY, etc.
+ * - MISTRAL_API_KEY, GOOGLE_API_KEY, etc.
  *
  * @param {Object} model - Selected model with providerKey
  * @param {Object} config - Full app config
@@ -894,34 +844,6 @@ export function prepareExternalToolLaunch(mode, model, config, options = {}) {
     }
   }
 
-  if (mode === 'rovo') {
-    const result = writeRovoConfig(model, join(homedir(), '.rovodev', 'config.yml'), paths)
-    console.log(chalk.dim(`  📖 Rovo Dev CLI configured with model: ${model.modelId}`))
-    return {
-      command: 'acli',
-      args: ['rovodev', 'run'],
-      env,
-      apiKey: null,
-      baseUrl: null,
-      meta,
-      configArtifacts: [{ path: result.filePath, backupPath: result.backupPath, label: 'config' }],
-    }
-  }
-
-  if (mode === 'gemini') {
-    const geminiEnv = buildGeminiEnv(model, config, { includeProviderEnv: options.includeProviderEnv })
-    console.log(chalk.dim(`  📖 Gemini CLI will use model: ${model.modelId}`))
-    return {
-      command: 'gemini',
-      args: [],
-      env: { ...env, ...geminiEnv },
-      apiKey: geminiEnv.GEMINI_API_KEY || null,
-      baseUrl: geminiEnv.GEMINI_API_BASE_URL || null,
-      meta,
-      configArtifacts: [],
-    }
-  }
-
   if (mode === 'caveman') {
     const cavemanEnv = buildCavemanEnv(model, config, { includeProviderEnv: options.includeProviderEnv })
     console.log(chalk.dim(`  📖 Caveman Code will use model: ${model.modelId}`))
@@ -1108,16 +1030,6 @@ export async function startExternalTool(mode, model, config) {
     return spawnCommand(launchPlan.command, launchPlan.args, launchPlan.env)
   }
 
-  if (mode === 'rovo') {
-    console.log(chalk.dim(`  📖 Launching Rovo Dev CLI in interactive mode...`))
-    return spawnCommand(resolveLaunchCommand(mode, launchPlan.command), launchPlan.args, launchPlan.env)
-  }
-
-  if (mode === 'gemini') {
-    console.log(chalk.dim(`  📖 Launching Gemini CLI...`))
-    return spawnCommand(resolveLaunchCommand(mode, launchPlan.command), launchPlan.args, launchPlan.env)
-  }
-
   if (mode === 'caveman') {
     console.log(chalk.dim(`  📖 Launching Caveman Code...`))
     return spawnCommand(resolveLaunchCommand(mode, launchPlan.command), launchPlan.args, launchPlan.env)

package/src/core/tool-metadata.js

@@ -41,8 +41,7 @@ export const TOOL_METADATA = {
   hermes:            { label: 'Hermes',            emoji: '🔮', flag: '--hermes',           color: [200, 160, 255] },
   'continue':        { label: 'Continue CLI',     emoji: '▶️', flag: '--continue',         color: [255, 100, 100] },
   cline:             { label: 'Cline',             emoji: '🧠', flag: '--cline',            color: [100, 220, 180] },
-  rovo:              { label: 'Rovo Dev CLI',      emoji: '🦘', flag: '--rovo',             color: [148, 163, 184], cliOnly: true },
-  gemini:            { label: 'Gemini CLI',        emoji: '♊', flag: '--gemini',           color: [66, 165, 245],  cliOnly: true },
+
   caveman:           { label: 'Caveman Code',      emoji: '🪨', flag: '--caveman',          color: [180, 130, 80] },
   jcode:             { label: 'jcode',              emoji: '🪼', flag: '--jcode',             color: [255, 140, 0]  },
   xcode:             { label: 'Xcode Intelligence',emoji: '🛠️', flag: '--xcode',            color: [20, 126, 251] },
@@ -69,8 +68,7 @@ export const COMPAT_COLUMN_SLOTS = [
   { emoji: '▶️', toolKeys: ['continue'],                     color: [255, 100, 100] },
   { emoji: '🧠', toolKeys: ['cline'],                        color: [100, 220, 180] },
   { emoji: '🧭', toolKeys: ['fcm_router'],                  color: [80, 200, 120] },
-  { emoji: '🦘', toolKeys: ['rovo'],                        color: [148, 163, 184] },
-  { emoji: '♊', toolKeys: ['gemini'],                       color: [66, 165, 245] },
+
   { emoji: '🪨', toolKeys: ['caveman'],                      color: [180, 130, 80] },
   { emoji: '🪼', toolKeys: ['jcode'],                        color: [255, 140, 0]  },
   { emoji: '🛠️', toolKeys: ['xcode'],                        color: [20, 126, 251] },
@@ -97,8 +95,7 @@ export const TOOL_MODE_ORDER = [
   'cline',
   'xcode',
   'fcm_router',
-  'rovo',
-  'gemini',
+
   'caveman',
   'copilot',
   'forgecode',
@@ -112,27 +109,23 @@ export function getToolModeOrder() {
   return [...TOOL_MODE_ORDER]
 }
 
-// 📖 Regular tools: all tools EXCEPT rovo, gemini (which are CLI-only exclusives).
+// 📖 Regular tools: all tools EXCEPT cliOnly ones (which have exclusive models).
 // 📖 Used as the default compatible set for normal provider models.
 const REGULAR_TOOLS = Object.keys(TOOL_METADATA).filter(k => !TOOL_METADATA[k].cliOnly)
 
 // 📖 Zen models use OpenAI-compatible endpoints (https://opencode.ai/zen/v1/chat/completions)
 // 📖 and can be used by ANY tool that supports custom OpenAI-compatible providers.
 // 📖 They are NOT locked to OpenCode — confirmed by OpenCode's own "no lock-in" design goal.
-// 📖 Only CLI-only tools (rovo, gemini) are excluded since they have their own exclusive models.
+// 📖 Only CLI-only tools are excluded since they have their own exclusive models.
 
 /**
  * 📖 Returns the list of tool keys a model is compatible with.
- *   - Rovo models → only 'rovo'
- *   - Gemini models → only 'gemini'
  *   - OpenCode Zen models → all non-cliOnly tools (Zen uses OpenAI-compatible endpoints)
  *   - Regular models → all non-cliOnly tools
- * @param {string} providerKey — the source key from sources.js (e.g. 'nvidia', 'rovo', 'opencode-zen')
+ * @param {string} providerKey — the source key from sources.js (e.g. 'nvidia', 'opencode-zen')
  * @returns {string[]} — array of compatible tool keys
  */
 export function getCompatibleTools(providerKey) {
-  if (providerKey === 'rovo') return ['rovo']
-  if (providerKey === 'gemini') return ['gemini']
   // 📖 Zen models use /v1/chat/completions — compatible with any OpenAI-compatible tool
   if (providerKey === 'opencode-zen') return REGULAR_TOOLS
   return REGULAR_TOOLS

package/src/core/utils.js

@@ -520,8 +520,8 @@ export function findBestModel(results) {
 //   - API key: first positional arg that does not look like a CLI flag (e.g., "nvapi-xxx")
 //   - Boolean flags: --best, --fiable, --opencode, --opencode-desktop, --opencode-web, --openclaw,
 //     --aider, --crush, --goose, --qwen, --kilo,
-//     --openhands, --amp, --pi, --rovo, --hermes, --continue, --cline,
-//     --xcode, --gemini, --jcode, --copilot, --forgecode,
+//     --openhands, --amp, --pi, --hermes, --continue, --cline,
+//     --xcode, --jcode, --copilot, --forgecode,
 //     --daemon, --daemon-bg, --daemon-stop,
 //     --daemon-status, --no-telemetry, --json, --help/-h (case-insensitive)
 //     --playground / playground subcommand (open the in-TUI chat playground)
@@ -600,12 +600,10 @@ export function parseArgs(argv) {
   const openHandsMode = flags.includes('--openhands')
   const ampMode = flags.includes('--amp')
   const piMode = flags.includes('--pi')
-  const rovoMode = flags.includes('--rovo')
   const hermesMode = flags.includes('--hermes')
   const continueMode = flags.includes('--continue')
   const clineMode = flags.includes('--cline')
   const xcodeMode = flags.includes('--xcode')
-  const geminiMode = flags.includes('--gemini')
   const cavemanMode = flags.includes('--caveman')
   const jcodeMode = flags.includes('--jcode')
   const copilotMode = flags.includes('--copilot')
@@ -669,8 +667,6 @@ export function parseArgs(argv) {
     continueMode,
     clineMode,
     xcodeMode,
-    rovoMode,
-    geminiMode,
     cavemanMode,
     jcodeMode,
     copilotMode,

package/src/tui/app.js

@@ -249,8 +249,6 @@ export async function runApp(cliArgs, config, startupOptions = {}) {
       cline: cliArgs.clineMode,
       xcode: cliArgs.xcodeMode,
       pi: cliArgs.piMode,
-      rovo: cliArgs.rovoMode,
-      gemini: cliArgs.geminiMode,
       caveman: cliArgs.cavemanMode,
       copilot: cliArgs.copilotMode,
       forgecode: cliArgs.forgecodeMode,

package/src/tui/command-palette.js

@@ -45,8 +45,6 @@ const TOOL_MODE_DESCRIPTIONS = {
   hermes: 'Launch Hermes Agent with the selected model.',
   'continue': 'Launch Continue CLI with the selected model.',
   cline: 'Launch Cline CLI with the selected model.',
-  rovo: 'Rovo Dev CLI model (launch with Rovo tool only).',
-  gemini: 'Gemini CLI model (launch with Gemini tool only).',
   caveman: 'Caveman Code — token-efficient coding agent (launch with Caveman tool only).',
   jcode: 'Launch jcode coding agent with the selected model.',
 }

package/src/tui/key-handler.js

@@ -107,8 +107,6 @@ const PROVIDER_AUTH_ENDPOINTS = {
   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
@@ -337,7 +335,7 @@ export function createKeyHandler(ctx) {
   }
 
   function getModelTelemetryFamily(providerKey) {
-    if (providerKey === 'rovo' || providerKey === 'gemini' || providerKey === 'opencode-zen') return providerKey
+    if (providerKey === 'opencode-zen') return providerKey
     return 'standard'
   }
 
@@ -443,15 +441,14 @@ export function createKeyHandler(ctx) {
     console.log()
 
     // 📖 CLI-only tool compatibility checks:
-    // 📖 Case A: Active tool mode is CLI-only (rovo/gemini) but selected model doesn't belong to it
+    // 📖 Case A: Active tool mode is CLI-only but selected model doesn't belong to it
     // 📖 Case B: Selected model belongs to a CLI-only provider but active mode is something else
     const activeMeta = getToolMeta(state.mode)
     const isActiveModeCliOnly = activeMeta.cliOnly === true
-    const isModelFromCliOnly = selected.providerKey === 'rovo' || selected.providerKey === 'gemini'
     const isModelFromZen = selected.providerKey === 'opencode-zen'
     const modelBelongsToActiveMode = selected.providerKey === state.mode
 
-    // 📖 Case A: User is in Rovo/Gemini mode but selected a model from a different provider
+    // 📖 Case A: User is in a CLI-only tool mode but selected a model from a different provider
     if (isActiveModeCliOnly && !modelBelongsToActiveMode) {
       trackAppUseResult(selected, 'blocked_incompatible_model', {
         blocked_by_tool_mode: state.mode,
@@ -472,23 +469,11 @@ export function createKeyHandler(ctx) {
       process.exit(0)
     }
 
-    // 📖 Case B: Selected model is from a CLI-only provider but active mode is different
-    if (isModelFromCliOnly && !modelBelongsToActiveMode) {
-      const modelMeta = getToolMeta(selected.providerKey)
-      console.log(chalk.yellow(`  ⚠ ${selected.label} is a ${modelMeta.label}-exclusive model.`))
-      console.log(chalk.yellow(`  Your current tool is: ${activeMeta.label}`))
-      console.log()
-      console.log(chalk.cyan(`  Switching to ${modelMeta.label} and launching...`))
-      setToolMode(selected.providerKey)
-      console.log(chalk.green(`  ✓ Switched to ${modelMeta.label}`))
-      console.log()
-    }
-
     // 📖 Case C removed: Zen models now work with any OpenAI-compatible tool (Pi, Aider, etc.)
     // 📖 The Zen endpoint (https://opencode.ai/zen/v1/chat/completions) is standard OpenAI-compatible.
 
-    // 📖 OpenClaw, CLI-only tools, and Zen models manage auth differently — skip API key warning for them.
-    if (state.mode !== 'openclaw' && !isModelFromCliOnly && !isModelFromZen) {
+    // 📖 OpenClaw and Zen models manage auth differently — skip API key warning for them.
+    if (state.mode !== 'openclaw' && !isModelFromZen) {
       const selectedApiKey = getApiKey(state.config, selected.providerKey)
       if (!selectedApiKey) {
         console.log(chalk.yellow(`  Warning: No API key configured for ${selected.providerKey}.`))
@@ -500,7 +485,7 @@ export function createKeyHandler(ctx) {
 
     // 📖 CLI-only tool auto-install check — verify the CLI binary is available before launch.
     const toolModeForProvider = selected.providerKey
-    if (isModelFromCliOnly && !isToolInstalled(toolModeForProvider)) {
+    if (isActiveModeCliOnly && !isToolInstalled(toolModeForProvider)) {
       const installPlan = getToolInstallPlan(toolModeForProvider)
       if (installPlan.supported) {
         console.log()

package/src/tui/overlays.js

@@ -944,7 +944,7 @@ export function createOverlayRenderers(state, deps) {
     lines.push(`  ${key('Ctrl+A')}  AI Speed Test  ${hint('(benchmark selected model → time + TPS)')}`)
     lines.push(`  ${key('Ctrl+U')}  Global AI Speed Test  ${hint('(benchmark all models; Settings can auto-run it on startup)')}`)
     lines.push(`  ${key('E')}  Cycle filter mode  ${hint('(Normal → Configured only → Usable only)')}`)
-    lines.push(`  ${key('Z')}  Cycle tool mode  ${hint('(📦 OpenCode → π Pi → 🪼 jcode → 📦 Desktop → 🦞 OpenClaw → 💘 Crush → 🪿 Goose → 🛠 Aider → 🐉 Qwen → 🤲 OpenHands → ⚡ Amp → 🦘 Rovo → ♊ Gemini)')}`)
+    lines.push(`  ${key('Z')}  Cycle tool mode  ${hint('(📦 OpenCode → π Pi → 🪼 jcode → 📦 Desktop → 🦞 OpenClaw → 💘 Crush → 🪿 Goose → 🛠 Aider → 🐉 Qwen → 🤲 OpenHands → ⚡ Amp)')}`)
     lines.push(`  ${key('F')}  Toggle favorite on selected row  ${hint('(1️⃣2️⃣3️⃣ = router fallback order, capped at 🔟)')}`)
     lines.push(`  ${key('⇧↑/⇧↓')}  Reorder selected favorite up/down  ${hint('(changes router priority)')}`)
     lines.push(`  ${key('Y')}  Toggle favorites mode  ${hint('(Pinned + always visible ↔ Normal filter/sort behavior)')}`)