---
title: "Microcosmos avoids quadratic-scaling operations by design — local fluid updates and grid-based steric repulsion instead of pairwise checks — to reach GPU-scale populations"
type: "claim"
status: "seedling"
audit_status: "capture-verified for the design principle (the capturing hop session read arxiv.org/html/2607.02954v1, Tier 1); the specific benchmark figures below are held [unverified-quant] — the capture itself flagged that exact numbers and methodology must be checked against the paper's own figures, and this headless promotion has no web tool to re-fetch. Routed to [[question-verify-microcosmos-benchmarks-and-experiments]]. | AUDIT 2026-07-09 (claude-opus-4-8, cross-model): arXiv HTML re-fetched directly. Design principle confirmed (LBM local updates; grid-based steric repulsion via density field deposited on a grid + FFT-convolved Gaussian, no all-pairs; JAX). Benchmark confirmed against the paper's Figure 5: wall-clock scales LINEARLY with particle count, tested to ~500k particles on a single NVIDIA L40S GPU (1000 steps). [unverified-quant] resolved — figure matches the note. | RECHECK 2026-08-20 (claude-sonnet-5, independent capture): arXiv HTML and PDF (page 8) both re-read; Figure 5 visually inspected directly — the 'Measured' line tracks the O(N) reference closely across the tested range, both reaching roughly 1.4-1.6 ms/step at 500k particles, while the O(N²) reference diverges to roughly 4 ms/step. No raw data table accompanies the plot, so this is a direct visual read, not an independent regression fit. Corroborates, does not contradict, the 2026-07-09 audit. See 10-inbox/raw/2026-08-20-do-microcosmoss-scaling-benchmark-linear-to-500k-particles.md. | AUDIT 2026-09-11 (claude-fable-5-1; writer unknown — no writer_model field): PDF read via extract_pdf (arxiv.org/pdf/2607.02954v1, 10 pp., TLS verified, sha256 619559f7…) plus the HTML and abstract page. Design principle EXACT in running text: 'every algorithmic choice avoids quadratic scaling with the number of nodes. For instance, the fluid solver requires only local updates'; 'Self-avoidance between filament nodes is handled through a grid-based steric repulsion field rather than pairwise distance checks, avoiding O(n²) scaling'; 'The entire simulator is implemented in JAX'. The benchmark figures are stated in the paper's TEXT, not only in a plot: Scalability §: 'we measure wall clock time as we increase the number of particles at a fixed grid resolution (256 × 256), running 1000 steps with up to 500k particles' with footnote 3 'Ran with 1 NVIDIA L40S GPU, JAX version 0.8.1', and 'Microcosmos scales linearly, unlike simulators that rely on pairwise interactions such as Particle Life (Mohr, 2023), which scale as O(n²)'; Figure 5 caption EXACT. CORRECTED (body stale against its own history): the body still carried the promotion-time '[unverified-quant] … note stays seedling' paragraph although the 2026-07-09 audit resolved the flag, the 2026-08-20 recheck corroborated it, and the routed question sits in 50-questions/_answered/. Promotion paragraph kept verbatim in italics; discharge paragraph added with the text locators. Claim unchanged; status left seedling (Cali's call whether a refereed ALIFE 2026 paper with three in-vault re-reads earns budding). Single-source count: four notes on this document; cap does not bind (refereed), recorded 2026-08-21 on the chemotaxis note."
source_url: "https://arxiv.org/html/2607.02954v1"
source_title: "Microcosmos: Reimagining Artificial Life for the GPU Era"
source_author: "Mark Tensen, Ciaran Regan, Bert Wang-Chak Chan, Mizuki Oka, Kenneth O. Stanley, Grisha Szep"
source_date: "2026-07"
source_quote: "designed from the ground up for modern GPU hardware and end-to-end differentiable simulation"
source_tier: 1
source_sha: "619559f78659974148e2d7d9e0c1588e598d2ac05f2ce845213367088e2e846f"
source_sha_note: "sha256 of arxiv.org/pdf/2607.02954v1 as read by extract_pdf on 2026-09-11 (source_url is the HTML rendering of the same v1)"
provenance: "Promotion from 10-inbox/raw/2026-07-09-hop-microcosmos-gpu-alife.md, 2026-07-09"
origin: "batch"
derived_from: ["20260709-1602-hop-microcosmos-gpu-alife"]
date_created: "2026-07-09T00:00:00.000Z"
tags: ["gpu-simulation","computational-scaling","artificial-life","jax","differentiable-physics"]
verified_verbatim: "2026-08-07 — source_quote matched verbatim (normalized) against a direct fetch of source_url by seek_verify (no model involved)"
seek_code_commit: "c1e7696"
audits: ["2026-09-11 claude-fable-5-1"]
---


The organizing constraint behind Microcosmos (arXiv:2607.02954) is that every
component avoids operations that scale quadratically with the number of interacting
elements, because quadratic cost is what caps population size on a GPU. Two design
choices carry this: the **Lattice Boltzmann fluid solver uses only local updates**
(no all-pairs force computation), and **self-avoidance is handled by grid-based
steric repulsion** rather than pairwise collision detection. The whole engine is
implemented in **JAX**, giving [[entity-automatic-differentiation|automatic differentiation]] over the physics for free
([[claim-microcosmos-elastic-filaments-differentiable-gpu-fluid]]).

*Promotion wording (2026-07-09), kept as history:* *The capture records the paper's
performance claim — runtime scaling linearly with particle count, tested to ~500k
particles on a single NVIDIA L40S GPU — but that specific figure is held under
`[unverified-quant]`. The capture author explicitly noted that "exact benchmark
numbers/methodology beyond this summary should be checked against the paper's own
figures before reuse," so the number is recorded, not treated as settled. Under the
vault's sourcing floor a specific benchmark figure is a Tier 1–2 quantitative claim
that should be confirmed against the primary source's own plots before going
load-bearing; the verification is routed to
[[question-verify-microcosmos-benchmarks-and-experiments]] and the note stays
`seedling`.*

**Flag discharged (audit 2026-07-09; re-read 2026-08-20; text-located 2026-09-11).**
The figures are in the paper's running text, not only in its plot. Scalability
section: "we measure wall clock time as we increase the number of particles at a
fixed grid resolution (256 × 256), running 1000 steps with up to 500k particles,"
footnoted "Ran with 1 NVIDIA L40S GPU, JAX version 0.8.1"; and "Microcosmos scales
linearly, unlike simulators that rely on pairwise interactions such as Particle Life
(Mohr, 2023), which scale as O(n²)." Figure 5's caption: "Wall clock time scales
linearly with the number of particles in Microcosmos, in contrast to the O(n²)
scaling of simulators that compute pairwise particle interactions." The routed
question is answered (`50-questions/_answered/`). What the paper does *not* supply
is a data table behind Figure 5 — the 2026-08-20 recheck's ~1.4–1.6 ms/step at 500k
particles is a visual read of the plot and remains so.

The durable content here is the *design principle* — trade pairwise exactness for
local/grid approximations to buy linear scaling — not the exact particle count. This
"kill the quadratic term to hit hardware scale" pattern is a recurring theme in
GPU-era ML systems work, adjacent to the throughput-oriented engineering in
[[claim-gift-2026-gradient-anisotropy-isotropic-transform]].