Changelog

30 new models: Claude 5.5, GPT-6, Gemini 4

• NEW Added Claude Opus 5.5, Sonnet 5.5 and Haiku 5.5, GPT-6.1 Sol, GPT-6 Astra and GPT-6 Luna, Gemini 4 Argon and Gemini 3.8 Flash, Qwen3.8 Max Prime, Grok 4.7, DeepSeek V4.1 Flash and Mistral Large 4.
• NEW New Meta Muse lineup (Muse Wonder, Muse Spark 1.3 and Muse Spark Open) and image generation with Nano Banana 2.1.
• IMPROVE. Richer Free plan: Claude Haiku 5.5, GPT-6 Luna, DeepSeek V4.1 Flash and image generation with Nano Banana 2 Lite.
• IMPROVE. Model guide: a detailed page for every model in all 9 site languages, with a link to the latest version of each model.
• FIX Requests sent to some models could fail: this is fixed.
• FIX Removed 3 models discontinued by their providers (Claude Sonnet 4, Qwen3 Coder Plus, Qwen3.6 Max), replaced by their successors.
• FIX Model guide: display fixed in Portuguese, Russian, Chinese, Japanese and Arabic, prices corrected and missing pages added.

More reliable subscriptions

• FIX Your plan is activated as soon as you pay, including with a promo code or a different VAT rate.
• IMPROVE. If a payment fails, you keep your plan for 7 days while you settle it.

Beta launch

• NEW New recent models added to the routing engine.
• NEW Manual model execution: choose the AI model used for your request yourself.

New models & minor fixes

• NEW New recent models added to the routing engine.
• FIX Several minor bug fixes.

Prompt optimization & site improvements

• NEW Integration of the Promptly API for automatic prompt optimization before sending to the routing engine — better categorization accuracy, more relevant responses.
• NEW Changelog page: full update history accessible from the navigation.
• IMPROVE. Navigation links centered across all pages.
• IMPROVE. Switch to formal register (vouvoiement) across the entire site.

Secondary pages & full site experience

• NEW About page: project overview, problem statement, Mingo solution, and product values.
• NEW Roadmap page with 4 phases: Alpha, Beta, Public Launch, Maturity.
• NEW Dedicated Feedback page for alpha testers with rating (1–5), categories, and free text.
• NEW GDPR-compliant Privacy Policy page.
• NEW FAQ section on the homepage with 7 frequently asked questions in accordion format.
• IMPROVE. Orbital animation: added Zhipu AI (GLM-5) on ring 6.
• FIX Removed all emojis across the site, replaced with SVG icons.

Landing page & public presentation

• NEW Full landing page with orbital animation featuring 12 AI model providers.
• NEW Marquee of 40+ frontier models supported.
• NEW Light / dark theme with persistence and cross-tab synchronization.
• NEW Full FR / EN internationalization across all pages.
• NEW Alpha access modal with email capture.

Private Alpha launch

• NEW Intelligent routing engine with automatic detection across 7 categories: code, creative, research, reasoning, simple, conversation, multimodal.
• NEW integration 40+ frontier models (Claude, GPT, Gemini, DeepSeek, Grok, Mistral, Llama, Kimi, Qwen, GLM…).
• NEW Chat interface with 3 access tiers: Free, Basic, Pro.
• NEW Routing transparency: detected category, model used, and justification displayed for each response.
• NEW Inline feedback system: 1-to-5 rating directly in the chat.