@supaihq profile picture
Sup AI @supaihq

Sup AI: Multi-LLM Orchestration: Real-time synthesis, always cited verifiable sources, no hallucinations, persistent memory across 40 frontier models. Try Free

Joined June 2024
  • Tweets 142
  • Following 48
  • Followers 26
43 Photos and videos
Pinned Tweet
Sup AI
@supaihq
10 Dec 2025
New SOTA on Humanity's Last Exam (HLE) We have achieved 52.15% accuracy on the world's hardest open-source AI reasoning test, setting a new benchmark record. Sup AI is now outperforming every individual frontier model, including Gemini 3 Pro Preview and GPT-5 Pro. Our lead over the next best model? 7.49 points. Check the full evaluation & code: github.com/supaihq/hle/blob/… #AI #MachineLearning #HLE #SupAI
3
4
8
1,009
Sup AI is live on @ProductHunt 🚀 "Which AI model is the best?" Wrong question. The best model isn't a model. It's an orchestra. Sup AI runs 9 frontier models in parallel and synthesizes their answers→ 52.15% on HLE benchmark (without the help of tools). → Multi-model consensus (up to 9 models) → Ensemble RAG with live web your files → Every claim cited $10 free credit to start 20% off with code: PRODUCTHUNT Links below 👇
3
1
3
162
Try Sup AI: sup.ai Support us on Product Hunt: producthunt.com/products/sup…

A letter from Ken

A letter from Ken Mueller on what's changing at Sup AI, and what I'm building next.

sup.ai
1
2
21
We just launched Sup AI on @ProductHunt! We combine multiple AI models and use confidence scoring to give better answers with fewer hallucinations. #1 on Humanity's Last Exam: 52.15%. Beating every individual model. $10 starter credit to try it, and 20% off your first month with code "PRODUCTHUNT" producthunt.com/products/sup…

Sup AI: AI ensemble that scored #1 on Humanity's Last Exam | Product Hunt

Every LLM hallucinates. They just don't hallucinate the same things. Sup AI runs multiple LLMs (out of 339) in parallel, then synthesizes answers by measuring confidence on every segment. High...

producthunt.com
1
4
44
Love seeing @Perplexity ship Model Council. Multi-model is the right direction. At Sup AI, we've pushed this further: 9-model ensembles segment-level confidence scoring (logprob signals across every claim). Text can lie. A model can sound 100% confident while hallucinating. The math doesn't lie. Result: 52.15% HLE (SOTA) 3 questions solved where ALL 9 individual models failed. The future isn't "which model is best." It's "what does each model know vs. what is it guessing?"
A screenshot of the Sup AI "Expert Mode" dashboard showing a list of AI models. Each model is labeled with an effort level (Medium or Low) and a logprob percentage. The interface highlights 9 frontier models including Claude Sonnet 4.5 and GPT-5.2 Pro with high logprobs, while noting that the current synthesis is being performed using Claude Opus 4.5.

ALT A screenshot of the Sup AI "Expert Mode" dashboard showing a list of AI models. Each model is labeled with an effort level (Medium or Low) and a logprob percentage. The interface highlights 9 frontier models including Claude Sonnet 4.5 and GPT-5.2 Pro with high logprobs, while noting that the current synthesis is being performed using Claude Opus 4.5.

Perplexity
@perplexity_ai
Feb 5
Introducing Model Council in Perplexity. Run three frontier models at once, compare outputs, and get a more accurate, higher‑confidence answer. Available now on web only for Perplexity Max subscribers.
0:42
1
141
Sup AI
@supaihq
Jan 30
This is exactly right. And it compounds with model diversity. At Sup AI: 5 prompt variations × 9 frontier models = 45 reasoning paths cross-validated before synthesis. Single prompt on single model = leaving 90% of accuracy gains on the table. My friend Gary Gurevich built a "hyperplane metaprompt" that automates the prompt side: generates 5 non-overlapping angles, predicts objections, synthesizes with traceability. Full template 👇
God of Prompt
@godofprompt
Jan 29
Stanford researchers just published a prompting technique that makes today’s LLMs behave like better versions of themselves. It’s called “prompt ensembling” and it runs 5 variations of the same prompt, then merges the outputs. Here’s how it works 👇
1
1
110
Sup AI
@supaihq
Jan 30
Gary's Hyperplane Method: "Generate a metaprompt to restate any prompt 4 ways (sharpening, scope-widening, cross-domain). Each restatement's center of mass overlays the original but extends in NON-OVERLAPPING directions. Answer all 5. Predict my objections. Answer those. Synthesize with full traceability." [your prompt]
1
44
Sup AI
@supaihq
Jan 30
Run this through 9 models in parallel and you get 45-path reasoning automatically. Diversity beats perfection. Every time.
37
Sup AI
@supaihq
Jan 28
Unpopular opinion: The AI model race is a distraction. See this tug-of-war? 👇 9 AI models vs. 1 "best" model. The crowd wins. Every time. No single LLM excels at everything: Claude crushes analysis, GPT-5 dominates creative, Gemini nails structured data. Orchestration intelligently routes each task to the RIGHT specialist. Sup AI proved it: 52.15% on Humanity's Last Exam, beating Gemini 3 Pro by 7.5 points. The companies winning in 2026 won't have the "best" model. They'll be the ones who stopped picking sides. Does orchestration become a first-class category this year? 👇 #AI #AIOrchestration #MultiModel
2
69
Sup AI
@supaihq
Jan 23
Microsoft CEO Satya Nadella just confirmed the Sup AI thesis: "Assigning roles to models and orchestrating them gets better results than any single frontier model." We’ve built the engine to prove it. • 52.15% accuracy • 7.4 percentage points vs. single models • Available today Stop waiting for the next GPT. Start orchestrating. 🎯
0:40
2
74
Sup AI
@supaihq
Jan 22
AI agents don't fail like chatbots… AI agents fail like software in production. One bad action breaks trust. @usevemly AI employees close tickets and update CRMs in live systems. Early on: too confident, too many errors. Fix: Sup AI as decision layer → Multiple models propose actions → Only executes on high consensus confidence → Otherwise: blocked or escalated Results: • 93% fewer incorrect tool calls * 41% faster resolution * 100% enterprise approval Full case study: sup.ai/case-studies/vemly Autonomy you can actually trust. #AgenticAI #EnterpriseAI
1
32
Sup AI
@supaihq
Jan 21
☑️ Pro Mode → Expert Mode ☑️ Orchestrator now auto-picks thinking effort per model = massive cost savings fixes slow GPT-5.2 Pro ☑️ Advanced model selector with per-model controls ☑️ Timestamps generation times on all messages
Main Section: Expert Mode Below the navigation bar, a header reads "Expert Mode" with a small circuit-board icon and a downward arrow. Underneath is a vertical list of nine AI models. Each entry shows a green checkmark (indicating it's selected), the model's logo, its name, a colored priority tag, and a percentage score: GPT-5.2 Pro — Yellow "Medium" tag — 54.98% Claude Sonnet 4.5 — Green "Low" tag — 73.11% Gemini 3 Pro — Green "Low" tag — 82.49% Grok 4.1 Fast Reasoning — Green "Low" tag — 71.09% Claude Haiku 4.5 — Blue "Extra Low" tag — 71.09% DeepSeek V3.2 Thinking — Green "Low" tag — 52.5% GPT-5.2 — Green "Low" tag — 66.82% Qwen3 Max — Green "Low" tag — 56.22% Mistral Large — Blue "Extra Low" tag — 84.55% Footer A thin horizontal line separates the list from the footer. The footer displays a green checkmark followed by: "Sup AI Synthesis using Claude Opus 4.5." Visual Style Summary The interface uses a dark grey/black background with white and light grey text.

ALT Main Section: Expert Mode Below the navigation bar, a header reads "Expert Mode" with a small circuit-board icon and a downward arrow. Underneath is a vertical list of nine AI models. Each entry shows a green checkmark (indicating it's selected), the model's logo, its name, a colored priority tag, and a percentage score: GPT-5.2 Pro — Yellow "Medium" tag — 54.98% Claude Sonnet 4.5 — Green "Low" tag — 73.11% Gemini 3 Pro — Green "Low" tag — 82.49% Grok 4.1 Fast Reasoning — Green "Low" tag — 71.09% Claude Haiku 4.5 — Blue "Extra Low" tag — 71.09% DeepSeek V3.2 Thinking — Green "Low" tag — 52.5% GPT-5.2 — Green "Low" tag — 66.82% Qwen3 Max — Green "Low" tag — 56.22% Mistral Large — Blue "Extra Low" tag — 84.55% Footer A thin horizontal line separates the list from the footer. The footer displays a green checkmark followed by: "Sup AI Synthesis using Claude Opus 4.5." Visual Style Summary The interface uses a dark grey/black background with white and light grey text.

52
Sup AI
@supaihq
Jan 16
Sup AI memory just leveled up We upgraded from Voyage Multimodal 3 → 3.5 with @VoyageAI * Best-in-class multimodal RAG * More accurate chat memories * Hyper-personalized answers * Everything becomes permanent knowledge️ #SupAI #VoyageAI #Multimodal #RAG
1
44
Sup AI
@supaihq
Jan 14
Sup AI Chrome Extension is live Your address bar → direct access to frontier models with forced citations. → Default search goes to Sup AI → !g for instant Google fallback → mode=fast / thinking / deep-thinking / pro → models=gemini-3-flash or models=qwen3-max,gemini-3-flash → Zero permissions. Zero data collection. chromewebstore.google.com/de…
3
91
Sup AI
@supaihq
Jan 12
1/ AI just solved an Erdős problem confirmed by @terencetao GPT-5.2 cracked Problem #728, a conjecture unsolved for decades. But the breakthrough isn't "one smart model." It's the architecture.
1
1
3
460
Sup AI
@supaihq
Jan 12
2/ The solution required ORCHESTRATION: • GPT-5.2 generated the proof (intuition) @sama • Harmonic's Aristotle verified it in Lean (rigor) @vladtenev • Human feedback refined the approach @terencetao This is constructive synthesis in action.
1
3
58
Sup AI
@supaihq
Jan 12
3/ At Sup AI, we've seen this pattern work. Our multi-model orchestration scored 52.15% on Humanity's Last Exam: 7.49 points above any single frontier model. The future isn't bigger models. It's smarter systems.
3
50
Sup AI whitepaper is live on the methodology behind 52.15% on HLE: • 3 correct answers synthesized when EVERY model failed • Grok 4 (29%) uniquely solved 16 Qs vs GPT-5 Pro's 9 (40%) • Low correlation pairs >high accuracy pairs • 58.44% theoretical ceiling w/ models • 42% Qs unsolved by ANY model • Full methodology, IQ curves, correlation matrices: sup.ai/research/hle-white-pa… #AI #MachineLearning #OpenSource #AIResearch #EnsembleAI #AIOrchestration #HLE
0:14
2
3
473
Sup AI now accepts virtually ANY file: → Images (JPEG, PNG, GIF, WEBP, HEIC, SVG) → Office (Word, Excel, PowerPoint) → Dev (Jupyter, CSV, code, ZIP) → Docs (PDF, EPUB, text)
Sup AI infographic showing supported file formats: Images (JPG, PNG, WEBP, HEIC, SVG), Office (DOCX, XLSX, PPTX), Dev (IPYNB, CSV, CODE, ZIP), and Docs (PDF, EPUB, RTF) connected to central Sup AI logo labeled "The Universal Context Layer" on dark navy background with neon circuit lines.

ALT Sup AI infographic showing supported file formats: Images (JPG, PNG, WEBP, HEIC, SVG), Office (DOCX, XLSX, PPTX), Dev (IPYNB, CSV, CODE, ZIP), and Docs (PDF, EPUB, RTF) connected to central Sup AI logo labeled "The Universal Context Layer" on dark navy background with neon circuit lines.

1
67
Load more