Yeeeeezus Intelligence — Inference Platform
Frontier models.
One route.
This is the static companion of the live platform: semantic search over the verified aisearch index runs MiniLM entirely inside this page, and the full 21-item triage table lives below. The live deployment adds streaming chat with Llama-3.3-70B, DeepSeek-V4, Qwen3.8-2.4T and 130+ more models through one route, plus image generation.
Semantic Search
Rank the index. In this tab.
all-MiniLM-L6-v2 (384-dim, quantized) downloads once and embeds every query locally — no text leaves your browser. Query with natural language: “open model for image generation”, “robot control”, “personal agent with a browser”.
booting inference runtime…
The aisearch index
Every item from the video, verified.
All 21 items from the aisearch video (2026-09-27) probed on 2026-09-30 against live GitHub, Hugging Face and web APIs. Access labels reflect what the APIs returned — not what the video claimed.
| Item | Class | Access | Verified evidence |
|---|
// closed-API items (GPT-6, Grok 4.7, Opus 5.5, Gemini 3.8 TTS…) cannot be upgraded by third parties —
// the honest move is better inputs and open counterparts, which is what the live platform routes to.
Catalog
Curated servable models.
Each of these was verified callable through the HF Inference Providers router on this account. The live deployment streams all of them; here they are linked to their weight pages.