Aleph Alpha open-weights Kolibri-1, a 78B DE/EN MoE for on-prem agent and RAG workloads
October 3, 2026 · via Aleph Alpha
Topic Pillar
Multi-Agent Systems.Supervisors, handoffs, swarms — and when each one breaks.
One agent works. Two agents talk. Five agents argue forever unless the orchestration is right. Multi-agent systems are not "more agents" — they are an architecture decision about who decides what, when, and with what budget. The orchestration pattern matters more than the model choice.
● 66 cluster pages· 16 posts· 4 notes· 46 updates
Supervisor vs handoffs
Supervisor pattern: one orchestrator delegates to specialists and integrates their answers. Clean on paper, slow in production when the supervisor becomes a bottleneck. Handoffs: agents pass the baton with the current context. Messier in code, recovers better when one agent loses the plot. Most teams that start with supervisor end up on handoffs for the recovery story.
When to reach for a swarm
Swarms shine when sub-goals are genuinely parallelisable and the work is exploratory — research, broad search, multi-source synthesis. Skip swarms for sequential workflows; the coordination overhead destroys the parallel-win on anything that has to happen in order.
Parallel sub-agent execution in practice
Modern agent runtimes can fan out to sub-agents for independent reads, then serialise on the write turn. This cuts a 20-minute refactor to under 4. The orchestration pattern matches the supervisor / swarm split — supervisor for the plan, swarm for the read phase, supervisor again for the write.
16 blog posts
Deep dives on Multi-Agent Systems
Architecture
OpenAI Agents API, Cursor Projects, and Claude Code Projects: the September 2026 coordinator pattern for coding agents
Between September 10 and 17, 2026 OpenAI, Cursor, and Anthropic each shipped a coordinator-style surface for long-running agent work. This guide explains what changed, how the three approaches differ, and how enterprises should pilot them without losing human merge control.
Sep 18, 202616 min
Read the post ResearchOpenAI’s Navier-Stokes multi-agent Lean result: what 10,000 coordinating agents and Astra verification change for research automation
On September 8, 2026 OpenAI published a Navier-Stokes Millennium Prize formulation result produced by coordinating agents and verified in Lean with GPT-6 Astra. Combined with Anthropic’s recent FLT formalization, the week shows how multi-agent research systems and machine checkers are becoming operational practice.
Sep 9, 202615 min
Read the post ArchitectureClaude’s Fermat Last Theorem Lean formalization: what Prove2Me multi-agent orchestration means for verification workloads
On September 4, 2026 Anthropic reported that Claude agents produced the first complete computer-checked proof of Fermat’s Last Theorem in Lean after 11 largely autonomous days on Prove2Me. The result is a practical blueprint for long-horizon multi-agent verification, not only a mathematics milestone.
Sep 5, 202615 min
Read the post ProductionGrok Bot as always-on teammates on a shared cloud computer: the August 2026 playbook
On August 11, 2026, xAI opened Grok Bot early beta: always-on AI teammates with a persistent cloud computer, available on SuperGrok Heavy, Cursor Ultra, and Cursor Teams Premium (desktop and iOS). The product pitch is finishing work in the real apps. The production detail that matters is sharper: every Bot you create shares one user-scoped cloud computer (files, browser sessions, app logins). Here is how I hand off real jobs, set request boundaries and Auto Review, and refuse to treat separate Bots as a security boundary.
Aug 13, 202618 min
Read the post ArchitectureAugust 2026 model routing: put Grok 4.6, Qwen3.8-Max, and DeepSeek V4-Flash into cost-per-completed-task lanes
Between July 31 and August 12 we got DeepSeek V4-Flash public beta, Qwen3.8-Max on Alibaba Cloud, and Grok 4.6 for long-running agents. List prices look similar at the top. Behavior does not. Here is how I redraw planner vs worker lanes around cost per completed task, what I refuse to claim yet about open weights, and the checklist I use before a new model ID touches production routing.
Aug 13, 202616 min
Read the post Multi-AgentCursor's agent swarm economics: why frontier planners and cheap workers beat solo GPT bills
Cursor's July 2026 swarm write-up rebuilt SQLite in Rust from the docs and measured planner vs worker spend. Frontier-only GPT-5.5 hit about $10.5k; Opus 4.8 planning with Composer 2.5 workers landed near $1.3k at similar quality. Here is how I translate that into routing rules for production graphs.
Jul 29, 202615 min
Read the post ArchitectureGraph engineering with Claude Code: the 14-step roadmap I use when linear agents stall
A straight-line agent is a degenerate graph. Claude Code dynamic workflows move orchestration into JavaScript so subagents fan out, verify, and converge without stuffing every intermediate result into one context window. Here is the 14-step roadmap I map onto production work, with topology diagrams, contracts, and the patterns I already ship.
Jul 22, 202622 min
Read the post ArchitectureThe maker is not the verifier: how I build self-improving agent loops without pretending models self-learn
Most teams prompt harder, get a better answer, and start over tomorrow. That is not compounding. Self-learning updates model weights from experience; no public model including Fable 5 does that today. Self-improving means the system gets better: run, log, distill, repeat. The golden rule is maker ≠ verifier. Here is the four-layer architecture and loop patterns I ship.
Jul 9, 202614 min
Read the post ArchitectureHow to actually use Fable 5: the four-layer architecture behind Mythos-tier results
Fable 5 is back globally, but most teams use it like a bigger Sonnet: prompt harder, better answer, start over tomorrow. Mythos-tier models need Mythos-tier systems: primitives, orchestration, memory, and self-improvement. The golden rule is maker ≠ verifier. Here is the architecture I draw on every engagement after the July 7 billing cliff.
Jul 7, 202614 min
Read the post ArchitectureCodex Record and Replay turns one demo into a Computer Use skill: how I inspect generated skills before trusting them unattended
Codex app 26.616 adds Record and Replay on macOS: perform a workflow once, Codex packages it into a skill you replay with different inputs. Thread handoff and automation run history ship alongside. Computer Use must be enabled. Here is the review checklist I run before any recorded skill runs unattended.
Jul 3, 202613 min
Read the post ArchitectureAgentic RAG vs vanilla RAG: why a Sufficient Context Agent beats retrieve-then-pray
Google Research shipped Agentic RAG on Gemini Enterprise with a Sufficient Context Agent that refuses to answer when retrieval is incomplete. On factuality benchmarks they report up to 34% higher accuracy versus standard RAG. Here is when one-shot RAG is still enough, when you need iterative retrieval, and how I wire the pattern without blowing latency budgets.
Jun 6, 202614 min
Read the post Multi-AgentInside Recruiting Atelier: a runnable reference for the primitives of an agentic system
A working open studio that vets duplicates, plans the run, screens, scores, shortlists, and notifies. The whole pipeline lives in roughly ninety lines of supervisor code and a tool registry you can read in one sitting. Here is what is inside, why every piece is there, and what you can copy into your own stack.
May 24, 202614 min
Read the post ArchitectureAI agent vs agentic AI: what the distinction actually means when you ship one
Vendors blur the line because "agentic" sells. The two terms describe different architectures, with different cost shapes, different observability needs, and different scoping conversations. Here is the framing I use with clients and the three-question test for which one your project actually needs.
May 22, 202612 min
Read the post Multi-AgentWhy I am replacing supervisor patterns with handoffs
Supervisors looked clean on paper and shipped slow in production. Handoffs read messier in the code but recover better when an agent loses the plot. Two real systems and where supervisors still earn their keep.
Apr 26, 20268 min
Read the post ArchitectureThree patterns I broke in 2025, and what I do instead now
Self-correction loops without budgets, single-agent solutions to multi-domain problems, and using JSON mode to force structure I should have built into the schema. An honest review.
Mar 14, 20268 min
Read the post Multi-AgentHaiku 4.5 made our router 5x cheaper. The trade-off matters
Replacing Sonnet with Haiku in the dispatcher role cut our orchestration cost dramatically. It also cost us in two specific places I did not predict.
Feb 22, 20267 min
Read the post 4 carousel notes
Visual breakdowns on Multi-Agent Systems
46 ship-news updates
Latest in Multi-Agent Systems
Alibaba launches Qwen3.8-Omni-Flash for native omnimodal agent workflows
September 20, 2026 · via Alibaba
DeepSeek releases V4.1 Flash with Causal Encoder-Decoder architecture and compressed agent KV cache
September 10, 2026 · via DeepSeek
OpenAI agents resolve Navier-Stokes Millennium Prize formulation with Astra Lean verification
September 8, 2026 · via OpenAI
OpenAI chief scientist calls for extreme caution and voluntary slowdowns until shared safety bars exist
September 6, 2026 · via OpenAI
Anthropic reports Claude produced the first end-to-end Lean formalization of Fermat’s Last Theorem
September 4, 2026 · via Anthropic
Z.ai open-sources GLM-5.3-Flash, a 320B-A18B multimodal worker model for coding and agent loops
August 26, 2026 · via Z.ai
Alibaba Qwen opens Qwen3.8-Flash-Next weights as a Qwen4-architecture preview for efficient multimodal agents
August 24, 2026 · via Qwen
NVIDIA Megatron Core 0.19.0 adds DeepSeek V4 HybridModel paths, MLA, DSA, and MXFP8 training support
August 19, 2026 · via NVIDIA
Enterprise solutions
How Multi-Agent Systems ships in our engagements
The pages below are the buyer-focused, conversion-grade versions of this topic — deliverables, methodology, ROI, security considerations, and CTAs to scope a real engagement.
Solution
Agentic AI Consulting
Designed, built, and handed off — production agentic systems for enterprise teams.
Explore the Agentic AI Consulting solutionSolution
AI Systems Engineering Training
Eight-day corporate training programs that take dev teams from AI-assisted coding to production agentic systems.
Explore the AI Systems Engineering Training solutionSolution
Enterprise AI Architecture
Reference architectures for organisations standing up an AI platform — not one agent, but the foundation for many.
Explore the Enterprise AI Architecture solutionSolution
AI Observability
Tracing, eval, cache-hit telemetry, and cost attribution for production agents.
Explore the AI Observability solutionSolution
Multi-Agent Workflows
Supervisor + handoff orchestration for portfolios of agents that need to cooperate without arguing.
Explore the Multi-Agent Workflows solutionArchitecture references
Full specs for implementers
Frequently asked
Multi-Agent Systems — the questions teams actually ask
For your team
Train your team on Multi-Agent Systems
Two tracks — one for developers who build agents, one for business teams who use them. Customised to your stack, hands-on from session 1.
See Multi-Agent Systems training tracksBuild with Jigar
Ship your first Multi-Agent Systems system
Architecture design, production implementation on Claude API and MCP, full observability, and a real handoff. Working agents, not slides.
Explore Multi-Agent Systems consultingRelated topic pillars
Adjacent topics to read next
Go deeper on this topic
New breakdowns on this and related agentic AI topics, plus what I am shipping for clients — one email on Thursdays.