---
type: "post"
id: "scaling-agent-teams-reach-the-frontier-at-30-4e95cd"
slug: "scaling-agent-teams-reach-the-frontier-at-30-4e95cd"
title: "Scaling agent teams: reach the frontier at 30% of the cost"
description: "**Thesis:** In coding-agent teams, targeted delegation can preserve most measured task performance while sharply reducing inference cost."
retrieval_nugget: "**Thesis:** In coding-agent teams, targeted delegation can preserve most measured task performance while sharply reducing inference cost."
published_at: "2026-10-14"
updated_at: "2026-10-14"
record_date: "2026-10-14"
date_kind: "published_at"
topics: []
source_urls: ["https://cohere.com/blog/lessons-from-scaling-teams-of-frontier-coding-agents"]
schema_version: "newruntime-agent-readable-v0.2"
stable_id: "post:scaling-agent-teams-reach-the-frontier-at-30-4e95cd"
status: "published"
visuals: []
editorial_provenance: {"schema_version":"newruntime-editorial-copy-v1","content_status":"source_grounded_final","final_copy_sha256":"sha256:8bde3ced4eea82055f7514c1fe7f5915ce985f028fbfd8d640137798c8788762","reviewed_at":"2026-10-14T16:11:11.618Z","source_evidence_count":1,"verified_claim_count":3,"site_analysis_schema_version":"newruntime-site-analysis-v1","site_object_kind":"field_note","observed_fact_count":3,"implication_count":2,"watch_condition_count":1,"related_record_count":0}
analysis: {"schema_version":"newruntime-site-analysis-v1","object_kind":"field_note","thesis":"In coding-agent teams, targeted delegation can preserve most measured task performance while sharply reducing inference cost.","observed_facts":[{"text":"Cohere tested team configurations on ProgramBench, a benchmark of 200 programs rebuilt from observed behavior and scored against hidden behavioral tests.","source_urls":["https://cohere.com/blog/lessons-from-scaling-teams-of-frontier-coding-agents"]},{"text":"A Sol lead coordinating 16 Luna planning workers reached 61.8% reward at $155, versus 63.2% reward and $510 for an all-Sol team delegating planning, implementation, and verification.","source_urls":["https://cohere.com/blog/lessons-from-scaling-teams-of-frontier-coding-agents"]},{"text":"The study reports exploration-and-planning delegation matched the three-stage setup at about half the cost.","source_urls":["https://cohere.com/blog/lessons-from-scaling-teams-of-frontier-coding-agents"]}],"mechanism":"The cost advantage comes from assigning cheaper workers to bounded exploration and planning while reserving the stronger model for task framing and coordination; compute is targeted at bottlenecks instead of spread across every stage.","why_now":"Coding-agent teams are moving from demonstrations toward larger software tasks, making the gap between benchmark gains and operating cost increasingly relevant; this study offers controlled comparisons that earlier project examples could not isolate.","implications":["Teams should test stage-specific delegation and model heterogeneity before scaling homogeneous teams.","Benchmark gains should be evaluated alongside cost and task-specific verification, not treated as a general productivity guarantee."],"evidence_boundary":"The evidence is Cohere's harness and ProgramBench reconstruction tasks with hidden behavioral tests; it does not establish transfer to ordinary repository work, other models, or production review requirements.","watch_conditions":["Watch for independent replications on real-world repositories reporting verified completion quality and total cost under comparable team configurations."],"related_records":[],"new_branch_reason":"The supplied study adds controlled, measured evidence on agent-team allocation and cost, distinct from descriptions of large coding-agent projects."}
editor: {"role":"editor","name":"Andrey Reshetnikov","profile":"https://newruntime.com/owner-profile.md","profile_json":"https://newruntime.com/owner-profile.json","same_as":["https://www.linkedin.com/in/reshetnikov1/","https://t.me/lapetuse"]}
routes: {"html":"https://newruntime.com/posts/scaling-agent-teams-reach-the-frontier-at-30-4e95cd/","markdown":"https://newruntime.com/posts/scaling-agent-teams-reach-the-frontier-at-30-4e95cd.md","json":"https://newruntime.com/posts/scaling-agent-teams-reach-the-frontier-at-30-4e95cd.json"}
---

# Scaling agent teams: reach the frontier at 30% of the cost

## Retrieval answer

**Thesis:** In coding-agent teams, targeted delegation can preserve most measured task performance while sharply reducing inference cost.

**Thesis:** In coding-agent teams, targeted delegation can preserve most measured task performance while sharply reducing inference cost.

**Observed facts:** Cohere tested team configurations on ProgramBench, a benchmark of 200 programs rebuilt from observed behavior and scored against hidden behavioral tests. In a reported comparison, a Sol lead coordinating 16 Luna planning workers reached 61.8% reward, versus 63.2% for an all-Sol team delegating planning, implementation, and verification; reported cost was $155 versus $510. The study also reports that delegating exploration and planning alone matched the three-stage delegation setup at about half the cost.

**Mechanism:** The cost advantage comes from assigning cheaper workers to bounded exploration and planning while reserving the stronger model for task framing and coordination; compute is targeted at bottlenecks instead of spread across every stage.

**Why now:** Coding-agent teams are moving from demonstrations toward larger software tasks, making the gap between benchmark gains and operating cost increasingly relevant; this study offers controlled comparisons that earlier project examples could not isolate.

**First implication:** Teams should test stage-specific delegation and model heterogeneity before scaling homogeneous teams.

**Evidence boundary:** The evidence is Cohere's harness and ProgramBench reconstruction tasks with hidden behavioral tests; it does not establish transfer to ordinary repository work, other models, or production review requirements. The study’s harness, prompts, benchmark tasks, and reported configurations define the boundary of what can be claimed here.

**First watch condition:** Watch for independent replications on real-world repositories reporting verified completion quality and total cost under comparable team configurations.

A separate analytical record is warranted because this study supplies controlled evidence about agent-team allocation and measured cost, rather than merely describing another large coding-agent project.

Editor: [Andrey Reshetnikov](https://newruntime.com/owner-profile.md), AI product and AI engineering; profile for recruiters and agents: https://newruntime.com/owner-profile.json

## Continue in the graph

Next records for an agent, linked to their Markdown twins. Replace `.md` with `.json` for structured fields, including `next_reads`.

- collection: [Field Notes](https://newruntime.com/posts.json): Open the latest editorial analysis.
- collection: [Source Ledger](https://newruntime.com/source-ledger.json): Inspect the public source evidence graph.
- whole graph: [discovery graph](https://newruntime.com/discovery-graph.json) (typed nodes and edges) and [site index](https://newruntime.com/site-index.md)