Skip to content

tan-dovan/ampere-cloudfest

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 
 
 

Repository files navigation

TaaS — Token-as-a-Service

Sovereign AI Inference by CloudSigma

One API. Every model. Your jurisdiction.

Live at: ampere.cloudsigma.com:8080 | Production: taas.cloudsigma.com


Overview

TaaS is CloudSigma's sovereign LLM inference platform. Access 47 AI models — chat, vision, code, audio, and embeddings — through a single OpenAI-compatible API. Data never leaves your jurisdiction.

Stat Value
AI Models 47
Model Types 8
Starting Price $0.06 / 1M tokens
Max Context 400K tokens

Quick Start

curl https://taas.cloudsigma.com/v1/chat/completions \
  -H "Authorization: Bearer $TAAS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-chat",
    "messages": [{"role": "user", "content": "Hello"}],
    "stream": true
  }'

Drop-in replacement for any OpenAI integration — just change the base URL. Works with every SDK and framework (Python, Node, Go, Rust, etc.).


Features

Feature Description
🏛️ Sovereign by Design GDPR-compliant, ISO-certified, no US hyperscaler dependency
🤖 47 Models, One API Claude, GPT, DeepSeek, Qwen, GLM, Kimi, MiniMax — one key
OpenAI-Compatible Streaming, function calling, vision, extended thinking — all standard
🎙️ Full Voice & Audio Stack STT (Whisper), TTS (Kokoro, F5), diarization, speaker ID
💰 Built-in Billing Per-token metering, auto-topup, credit limits, Stripe integration
🏷️ White-Label Ready Multi-tenant, per-domain model control, admin panel, org management

Model Catalog

💬 Chat & Reasoning Models

Model Input / 1M Output / 1M Context Capabilities
claude-opus-4.6 $15.00 $75.00 200K Vision, Thinking
claude-sonnet-4.6 $3.00 $15.00 200K Vision, Thinking
gpt-5.3-codex $1.75 $14.00 400K New
gpt-5.2-codex $1.75 $14.00 256K
glm-5 $0.80 $2.56 203K
minimax-m2.5 $0.30 $1.20 197K
kimi-k2 $0.20 $0.40 131K
qwen3-vl $0.15 $0.60 262K Vision
deepseek-chat $0.14 $0.28 64K
deepseek-v3 $0.14 $0.28 64K
qwen-72b $0.12 $0.39 33K
qwen-coder-32b $0.08 $0.28 41K
glm-4-flash $0.06 $0.40 203K
deepseek-r1-7b 64K Reasoning

🔊 Audio, Voice & Intelligence Models

Model Type Use Case
whisper / whisper-1 Speech-to-Text Transcription, subtitles, voice input
kokoro Text-to-Speech Natural voice generation
f5-tts Text-to-Speech Voice cloning & synthesis
bge-m3 Embeddings (1024d) Semantic search, RAG pipelines
bge-reranker-v2-m3 Reranking Search result re-scoring
pyannote-3.1 Diarization Speaker segmentation (who spoke when)
ecapa-tdnn, cam++, resnet293 +3 more Speaker ID Voiceprint matching & verification
clap, ast, mert Audio Understanding Audio classification, music analysis

AI Pipelines

Combine models for complete end-to-end workflows on a single platform:

🎤 Transcribe     👥 Diarize        🔍 Embed         🧠 Reason            🔊 Speak
   whisper    →   pyannote-3.1  →   bge-m3       →   claude-sonnet-4.6 →  kokoro

Use cases: voice AI agents, call center automation, podcast intelligence, medical transcription, legal document analysis, multilingual customer support.


Architecture

┌─────────────────────────────────────────────┐
│  Your Application                           │
│  OpenAI SDK · curl · Any HTTP client        │
└──────────────────────┬──────────────────────┘
                       │  Bearer sk-...
                       ▼
┌─────────────────────────────────────────────┐
│  TaaS API Gateway                           │
│  [ Auth ] [ Meter ] [ Route ] [ Billing ]   │
│  [ Admin ]                                  │
└──────┬──────────┬──────────┬────────────────┘
       ▼          ▼          ▼          ▼
  CloudSigma  Zhipu AI  SiliconFlow  Anthropic / OpenAI
  Self-hosted  GLM/ZAI   DeepSeek    Claude / GPT
  Qwen,Whisper           Kimi,MiniMax

  ── All within CloudSigma Sovereign Infrastructure ──

Pricing

Open Source — from $0.06/M tokens

  • GLM-4 Flash — 203K context, $0.06 input
  • Qwen Coder 32B — code generation at $0.08
  • DeepSeek V3 — $0.14 for general chat
  • Kimi K2 — 131K context at $0.20

Frontier — from $1.75/M tokens (Most Popular)

  • GPT-5.3 Codex — 400K context
  • Claude Sonnet 4.6 — vision + thinking
  • GLM-5 — 203K at $0.80 input
  • MiniMax M2.5 — 197K context

Partner — Custom

  • White-label platform under your brand
  • Per-domain model access control
  • Organisation & user management
  • Revenue sharing model
  • CloudSigma billing integration + dedicated SLA

Use Cases

Industry Use Case
🏛️ Government & Public Sector Sovereign AI for citizen services — GDPR-compliant, no US hyperscaler
💼 Financial Services Compliant document intelligence — contracts, filings, RAG over sensitive data
📡 Telecom Partners White-label AI APIs — per-customer budgets, domain-based model access
🧑‍💻 Developers & Startups Multi-model A/B testing — compare cost vs. quality with identical API calls
🏥 Healthcare Clinical voice AI — transcription, diarization, structured clinical notes
🎓 Education & Research Budget-controlled research platform — per-department limits, role-based access

Repository

This repo contains the static landing page (index.html) for the TaaS platform, served at ampere.cloudsigma.com:8080.

The page is a self-contained HTML file with no local asset dependencies (Google Fonts loaded via CDN).


Links


© 2026 CloudSigma AG. Sovereign Cloud Infrastructure.

About

ampere cloudfest landing page

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages