This page is built to give customers, partners, and enterprise teams continuous, real-time visibility into the operational health of Swarms services running in production environments.
As Swarms becomes core infrastructure for autonomous agent systems, transparency and reliability are non-negotiable.
Powered by @BetterStackHQ
Learn more ⬇️🧵
2 /
The Swarms Status Page will serve as the authoritative source for all platform-level communications, including:
• Live service availability and uptime metrics
• Scheduled maintenance and infrastructure upgrades
• Incident reports, historical timelines, and post-incident analysis
This ensures teams can confidently build, deploy, and operate at scale with full operational awareness.
Each service is continuously monitored to meet production and enterprise reliability expectations.
4 /
All future infrastructure updates, maintenance windows, and incident communications will be published through this status page as we continue expanding Swarms globally.
This is a foundational step toward operating Swarms as mission-critical infrastructure for the agent economy.
Our Latest update introduces new agent-level features, advanced agent routing strategies, deeper marketplace integration, and improved execution persistence.
The platform has been hardened for long-running, large-scale agent systems, while documentation and tutorials have been expanded to reflect real-world deployment patterns.
Learn more ⬇️🧵
2 /
[FEAT] Marketplace Prompt Fetching
Agents can now fetch and load prompts directly from the Swarms Marketplace, enabling dynamic prompt reuse, sharing, and versioned prompt orchestration across deployments.
Introduced support for adding agents in batches, improved node management, and new graph workflow examples leveraging rustworkx-style execution patterns.
Introducing Voice-Agents: An All-New Enterprise-Grade Voice Agent Framework 🗣️👾
Building voice-enabled agentic workflows just got easier, faster, and more reliable.
Voice-Agents is an all-new production-ready Python framework that provides seamless integration with multiple TTS/STT providers, real-time streaming, and everything you need to build conversational agentic assistants.
> Multi-provider support: OpenAI, ElevenLabs, and Groq
> Real-time streaming for low-latency agent interactions
> Production-ready with enterprise-grade logging, telemetry, and error-handling
Learn more ⬇️🧵
2 /
Multi-Provider TTS Support
Switch between providers effortlessly with a unified API. Whether you need OpenAI's natural voices, ElevenLabs' expressive options, or Groq's fast inference, Voice-Agents handles it all with consistent interfaces.
> 10+ OpenAI voices (alloy, nova, shimmer, and more)
> 30+ ElevenLabs voices with advanced voice control
> Unified stream_tts() function works across all providers
Built for agent-based systems that need low-latency audio streaming. Voice-Agents processes audio chunks as they arrive, enabling natural conversations without awkward pauses or delays.
> StreamingTTSCallback automatically speaks complete sentences from agent outputs
> Generator-based streaming for FastAPI and web applications
> Intelligent sentence detection for natural speech pauses
Through this program, we're helping select teams transform AI agent concepts into commercially viable, enterprise-scale businesses.
With access to our hardened infrastructure stack, funding, go-to-market execution, and integrated monetization through the Swarms Launchpad, Foundry teams are built to scale from day one.
Build. Deploy. Monetize.
Learn more now ⬇️🧵
2 /
The core problem in today’s AI and crypto landscape is economic, not technical.
Teams are building increasingly capable agents, but without distribution, pricing, or settlement infrastructure, those agents remain isolated tools dependent on speculative launches rather than real revenue.
3 /
Swarms solves the full Agent-to-Earn lifecycle from building intelligent agents to operating them as revenue-generating services.
Alongside robust orchestration infrastructure, Swarms introduces a global marketplace where agents are discovered, contracted, and paid from day one.
This is one of our largest and most comprehensive updates ever. We’ve spent the last 30 days working on this incredible release.
This update introduces powerful new multi-agent architectures such as LLMCouncil and DebateWithJudge, along with significant improvements to existing ones like GraphWorkflow and much more.
In addition, we’ve squashed 7 critical bugs, dramatically expanded test coverage for greater stability, shipped extensive documentation updates, and refreshed key dependencies including Pydantic and pytest.
The LLMCouncil class orchestrates multiple specialized LLM agents to collaboratively answer queries through a structured peer review and synthesis process.
Inspired by Andrej Karpathy's llm-council implementation, this architecture demonstrates how different models evaluate and rank each other's work, often selecting responses from other models as superior to their own.
From November 24 to December 2, we’re offering the biggest cost reductions of the year across our Swarms Cloud premium plans to celebrate Black Friday and Cyber Monday.
Our premium plans provide higher rate limits, global availability, exclusive multi-agent architectures, enterprise-grade security, and much more.
The Pro Plan is 75% off during Swarms Week. It’s our introductory production-tier offering and includes:
> Global edge routing
> Advanced multi-agent architectures such as GraphWorkflow and LLM Council
> 5× higher rate limits
> Full API telemetry dashboard
> 24/7 support
Normal Pricing: $19.99
Black Friday pricing: $4.99/month with code PRO75.
The Ultra Plan is built for high-volume, compliance-driven, and performance-critical environments. During Swarms Week, it’s 35% off and includes:
> 25× free-tier rate limits
> SOC 2 Type II compliance and audit logs
> Ultra-performance Rust-based agent runtime with near-zero overhead
> Advanced utilities such as agent version retrieval, marketplace agent access, and more
Normal Pricing: 99.99$
Black Friday pricing: $65/month with code SWARMS35.