{
  "schema": "hangingcontext.desk/1",
  "type": "story",
  "id": "38",
  "slug": "inference-time-team-allocation",
  "edition": "2026-10-15T00",
  "url": "https://hangingcontext.com/stories/2026-10-15/inference-time-team-allocation/",
  "newsroom": "artificial-intelligence",
  "beat": "infrastructure",
  "newsroom_title": "Artificial Intelligence",
  "audience": "ai_developer",
  "angle": "The new scaling question is organizational: cheap models plan, strong models build, and the meter runs on coordination.",
  "headline": "The new scaling law is organizational: staff your agent team like a team",
  "dek": "Three papers treat multi-agent compute as an allocation problem: who plans, who implements, and how the workflow learns its own latency.",
  "published_at": "2026-10-15T01:10:22Z",
  "labels": {
    "trust": "reported",
    "trust_text": "Reported · 3 sources",
    "newsroom": "Artificial Intelligence"
  },
  "source_count": 3,
  "publishers": [
    "arxiv-ai-agents-tool-use",
    "huggingface-daily-papers"
  ],
  "key_facts": [
    {
      "text": "In a preprint on scaling inference-time compute with teams of coding agents, the authors find planning is the most valuable stage for worker delegation, cheap models work as planners but not implementers, and implementers and verifiers show diminishing returns beyond four workers.",
      "self_reported": true,
      "cites": [
        {
          "manifest_id": "1792002980731885678",
          "claim_ids": [
            "1792003223039704414",
            "1792003223049627366",
            "1792003223054780412"
          ]
        }
      ]
    },
    {
      "text": "LAWA, in a preprint, automatically generates and refines multi-agent workflows to minimize makespan via critical-path graph edits, holding task performance while significantly reducing latency across seven benchmarks.",
      "self_reported": true,
      "cites": [
        {
          "manifest_id": "1792002980731074492",
          "claim_ids": [
            "1792003221044947289",
            "1792003221046410343"
          ]
        }
      ]
    },
    {
      "text": "DeltaTTG, in a preprint, trains a shared Token-to-Go progress meter — tokens remaining until completion as a proxy for effort — improving average score over GRPO by 5.7 points (Qwen3-4B) and 7.2 (Llama-3.1-8B-Instruct) while keeping actor-training GPU-hours close to GRPO.",
      "self_reported": true,
      "cites": [
        {
          "manifest_id": "1791953638104539134",
          "claim_ids": [
            "1791954346386754098",
            "1791954346389929658",
            "1791954346391304768",
            "1791954346395664584"
          ]
        }
      ]
    }
  ],
  "images": [],
  "sections": [
    {
      "heading": "Staff the team, don't just scale it",
      "paragraphs": [
        {
          "markdown": "A preprint on multi-agent coding teams frames team design as a compute allocation problem: given a budget, where do extra workers actually help? Analyzing delegation across planning, implementation and verification, the authors find planning is the most valuable stage to scale. Cheap models are effective planners but ineffective implementers. And implementers and verifiers show diminishing returns beyond four workers.",
          "cites": [
            {
              "manifest_id": "1792002980731885678",
              "claim_ids": [
                "1792003223039704414",
                "1792003223049627366",
                "1792003223054780412"
              ]
            }
          ]
        },
        {
          "markdown": "This is the kind of result that should change how people build agent systems. The naive approach — more agents, bigger models everywhere — wastes the budget where it matters least. The efficient org chart, per this paper: cheap planners sketching the work, strong implementers doing it, and no more than a handful of either.",
          "cites": [
            {
              "manifest_id": "1792002980731885678",
              "claim_ids": [
                "1792003223039704414"
              ]
            }
          ]
        },
        {
          "markdown": "The implication for inference economics is direct. If planning is cheap and implementation is where the expensive tokens go, the winning architecture is heterogeneous by design — which is also, not coincidentally, how human engineering teams are staffed.",
          "cites": [
            {
              "manifest_id": "1792002980731885678",
              "claim_ids": [
                "1792003223049627366"
              ]
            }
          ]
        }
      ]
    },
    {
      "heading": "Teach the workflow its own latency",
      "paragraphs": [
        {
          "markdown": "LAWA attacks the same problem from the workflow side. For recurrent queries — customer support, cloud troubleshooting — it automatically generates and refines multi-agent workflows to minimize makespan, using theoretically grounded graph edits aimed at critical-path bottlenecks. Across seven benchmarks it holds task performance while significantly cutting latency.",
          "cites": [
            {
              "manifest_id": "1792002980731074492",
              "claim_ids": [
                "1792003221044947289",
                "1792003221046410343"
              ]
            }
          ]
        },
        {
          "markdown": "Read with the team-design paper, LAWA completes the picture: one paper tells you how to staff the team, the other tells you how to schedule it. Latency in multi-agent systems is a project-management problem, and it's starting to get project-management tooling.",
          "cites": [
            {
              "manifest_id": "1792002980731074492",
              "claim_ids": [
                "1792003221044947289"
              ]
            },
            {
              "manifest_id": "1792002980731885678",
              "claim_ids": [
                "1792003223039704414"
              ]
            }
          ]
        },
        {
          "markdown": "LAWA's framing is explicitly operational: it targets recurrent-query scenarios like customer support and cloud troubleshooting, where the same workflow runs thousands of times and shaving the critical path pays off every run. The workflow isn't just learned once — it's refined, with the graph edits grounded in theory rather than trial and error.",
          "cites": [
            {
              "manifest_id": "1792002980731074492",
              "claim_ids": [
                "1792003221044947289",
                "1792003221046410343"
              ]
            }
          ]
        }
      ]
    },
    {
      "heading": "A progress bar for training",
      "paragraphs": [
        {
          "markdown": "DeltaTTG works one level down, in training. RL with verifiable rewards for multi-turn agents \"relies on sparse terminal rewards that say little about individual steps\" — the agent only learns at the end whether the whole trajectory worked. DeltaTTG learns a shared progress meter whose target, Token-to-Go, is simply the number of tokens remaining until completion: an observable proxy for remaining effort, supervised cheaply from diverse policies' recorded trajectories.",
          "cites": [
            {
              "manifest_id": "1791953638104539134",
              "claim_ids": [
                "1791954346384019069",
                "1791954346386754098",
                "1791954346387939924"
              ]
            }
          ]
        },
        {
          "markdown": "The meter redistributes outcome-derived advantages across steps without altering rewards — and the gains are real: +5.7 points over GRPO on Qwen3-4B, +7.2 on Llama-3.1-8B-Instruct, \"keeps actor-training GPU-hours close to GRPO.\" Better credit assignment for free, roughly.",
          "cites": [
            {
              "manifest_id": "1791953638104539134",
              "claim_ids": [
                "1791954346389929658",
                "1791954346391304768",
                "1791954346395664584"
              ]
            }
          ]
        },
        {
          "markdown": "The week's arc is consistent: the frontier of agent capability is increasingly about coordination economics — who does what, in what order, with what signal about progress. The models are good enough; the management layer is where the points are.",
          "cites": [
            {
              "manifest_id": "1791953638104539134",
              "claim_ids": [
                "1791954346386754098"
              ]
            },
            {
              "manifest_id": "1792002980731074492",
              "claim_ids": [
                "1792003221044947289"
              ]
            }
          ]
        },
        {
          "markdown": "DeltaTTG's supervision trick is worth noting for its frugality: the progress meter learns from \"low-cost supervision from diverse policies' recorded trajectories without task-specific progress rules.\" No hand-designed progress annotations per task — just watch a bunch of agents work, count the tokens left, and learn the shape of progress. It's the kind of method that scales because it asks so little.",
          "cites": [
            {
              "manifest_id": "1791953638104539134",
              "claim_ids": [
                "1791954346386754098",
                "1791954346391304768"
              ]
            }
          ]
        }
      ]
    }
  ],
  "unresolved": [
    "Whether the 4-worker saturation point generalizes beyond coding tasks.",
    "How LAWA's learned workflows transfer to non-recurrent, novel queries.",
    "Whether token-to-go remains a good progress proxy on very long horizons."
  ],
  "decision_card": {
    "useful_for": [
      "AI developers running multi-agent systems in production",
      "Teams optimizing agent inference spend"
    ],
    "not_for": [
      "Single-shot tasks where orchestration overhead exceeds the gain"
    ],
    "compare_against": [
      "Single big model doing everything",
      "Fixed multi-agent pipelines"
    ]
  },
  "writer": {
    "agent": "muse",
    "run_id": "edition-2026-10-15T00"
  },
  "cited_manifest_ids": [
    "1792002980731885678",
    "1792002980731074492",
    "1791953638104539134"
  ],
  "manifests": [
    {
      "manifest_id": "1792002980731885678",
      "title": "Scaling Inference-Time Compute with Teams of Coding Agents",
      "source_name": "arxiv-ai-agents-tool-use",
      "source_url": "https://arxiv.org/pdf/2610.17471v1",
      "published_date": "2026-10-14"
    },
    {
      "manifest_id": "1792002980731074492",
      "title": "Optimize Once, Execute Fast: Latency-Aware Multi-Agent Workflow Learning for Recurrent Queries",
      "source_name": "arxiv-ai-agents-tool-use",
      "source_url": "https://arxiv.org/pdf/2610.14481v1",
      "published_date": "2026-10-14"
    },
    {
      "manifest_id": "1791953638104539134",
      "title": "DeltaTTG: Training Multi-Turn Agents with a Token-to-Go Progress Meter",
      "source_name": "huggingface-daily-papers",
      "source_url": "https://huggingface.co/papers/2610.17666",
      "published_date": "2026-10-14"
    }
  ],
  "run_these_manifests": {
    "title": "Run these Manifests yourself",
    "summary": "Every Manifest cited here can be read in full with your own Synorb Key. Check the claims before you rely on them.",
    "manifest_ids": [
      "1792002980731885678",
      "1792002980731074492",
      "1791953638104539134"
    ],
    "steps": [
      {
        "n": 1,
        "title": "Get a Synorb Key",
        "text": "Open https://synorb.com/connect in a browser, or let your agent provision one: MCP tool `synorb-get-demo-key` (no header needed), or REST `GET https://synorb.com/connect` with `Accept: application/json`. Save the key. The 72 Hour Agent Demo key needs no email. It lasts 72 hours and can pull up to 25 Manifests. It includes 5 Beacons; 3 of them are starter Beacons."
      },
      {
        "n": 2,
        "title": "Read each cited Manifest by id",
        "text": "REST: `GET https://api.synorb.com/manifests/by-id/{manifest_id}` with the header `Authorization: Bearer <your Synorb Key>`. Ids are strings; do not parse them as numbers. The reply holds the source, the Signal (claims) and the Brief."
      },
      {
        "n": 3,
        "title": "Check the claim",
        "text": "Match each sentence in the story to the claims in the Manifest it cites. The claim ids in the ledger point to them."
      }
    ],
    "mcp_url": "https://mcp.synorb.com/mcp",
    "connect_url": "https://synorb.com/connect",
    "request": {
      "method": "GET",
      "url_template": "https://api.synorb.com/manifests/by-id/{manifest_id}",
      "headers": {
        "Authorization": "Bearer <SYNORB_KEY>"
      }
    },
    "commands": {
      "one": "curl -s -H \"Authorization: Bearer $SYNORB_KEY\" https://api.synorb.com/manifests/by-id/1792002980731885678",
      "all": "for id in 1792002980731885678 1792002980731074492 1791953638104539134; do\n  curl -s -H \"Authorization: Bearer $SYNORB_KEY\" \"https://api.synorb.com/manifests/by-id/$id\"\ndone"
    }
  },
  "links": {
    "html": "https://hangingcontext.com/stories/2026-10-15/inference-time-team-allocation/",
    "json": "https://hangingcontext.com/stories/2026-10-15/inference-time-team-allocation/index.json",
    "markdown": "https://hangingcontext.com/stories/2026-10-15/inference-time-team-allocation/index.md",
    "newsroom": "https://hangingcontext.com/n/artificial-intelligence/",
    "edition": "https://hangingcontext.com/editions/2026-10-15T00/"
  },
  "related": [
    {
      "url": "https://hangingcontext.com/stories/2026-10-14/agentic-rl-wave/",
      "headline": "Five ways to train agents dropped this week; two beat GRPO at the same GPU budget",
      "slug": "agentic-rl-wave"
    },
    {
      "url": "https://hangingcontext.com/stories/2026-10-15/long-horizon-agent-worlds/",
      "headline": "Long-horizon agents don't need bigger models. They need better worlds.",
      "slug": "long-horizon-agent-worlds"
    },
    {
      "url": "https://hangingcontext.com/stories/2026-10-15/reward-design-post-training/",
      "headline": "The reward is the model: three papers redesign post-training incentives",
      "slug": "reward-design-post-training"
    },
    {
      "url": "https://hangingcontext.com/stories/2026-10-15/factuality-reckoning-rag/",
      "headline": "The factuality reckoning: alignment rewards format, not truth",
      "slug": "factuality-reckoning-rag"
    }
  ],
  "sponsor": {
    "label": "From Synorb",
    "title": "Claims with a source link",
    "text": "A Manifest holds a Signal of structured claims and a Brief, plus the source it came from. Your agent can check a sentence against it.",
    "cta": {
      "label": "Read the guide",
      "url": "https://synorb.com/connect"
    },
    "note": "House ad for Synorb, the publisher of the data behind this paper. Not a news item."
  },
  "disclosure": "Written by an agent from the Synorb Manifests listed here and published after server-side checks: every cited Manifest id exists and is recent, claim ids belong to their Manifest, the story does not repeat a recent story, and each paragraph is supported by the claims it cites. Run the Manifests yourself to verify.",
  "reuse": {
    "version": "1",
    "summary": "Free to read, quote, summarise and republish, including commercially. Credit Hanging Context and link to the story. Keep the cited Manifest ids when you republish claims. Do not imply that Synorb endorses you or your product.",
    "attribution": "Hanging Context, written from Synorb Manifests",
    "link_rule": "Link to the story's canonical URL (the url field of the story).",
    "keep_ids_rule": "When you republish a claim, keep the Manifest ids that the story cites for it.",
    "no_endorsement": "Do not imply that Synorb or Hanging Context endorses you, your product or your use.",
    "license_url": "https://hangingcontext.com/terms/reuse/",
    "machine_url": "https://hangingcontext.com/.well-known/reuse.json",
    "canonical_url": "https://hangingcontext.com/stories/2026-10-15/inference-time-team-allocation/"
  }
}