chore(ci): bump docker/setup-buildx-action from 3 to 4

[dependabot]

Bumps docker/setup-buildx-action from 3 to 4.

Release notes

Sourced from docker/setup-buildx-action's releases.

v4.0.0

• Node 24 as default runtime (requires Actions Runner v2.327.1 or later) by @​crazy-max in docker/setup-buildx-action#483
• Remove deprecated inputs/outputs by @​crazy-max in docker/setup-buildx-action#464
• Switch to ESM and update config/test wiring by @​crazy-max in docker/setup-buildx-action#481
• Bump @​actions/core from 1.11.1 to 3.0.0 in docker/setup-buildx-action#475
• Bump @​docker/actions-toolkit from 0.63.0 to 0.79.0 in docker/setup-buildx-action#482 docker/setup-buildx-action#485
• Bump js-yaml from 4.1.0 to 4.1.1 in docker/setup-buildx-action#452
• Bump lodash from 4.17.21 to 4.17.23 in docker/setup-buildx-action#472
• Bump minimatch from 3.1.2 to 3.1.5 in docker/setup-buildx-action#480

Full Changelog: docker/setup-buildx-action@v3.12.0...v4.0.0

v3.12.0

• Deprecate install input by @​crazy-max in docker/setup-buildx-action#455
• Bump @​docker/actions-toolkit from 0.62.1 to 0.63.0 in docker/setup-buildx-action#434
• Bump brace-expansion from 1.1.11 to 1.1.12 in docker/setup-buildx-action#436
• Bump form-data from 2.5.1 to 2.5.5 in docker/setup-buildx-action#432
• Bump undici from 5.28.4 to 5.29.0 in docker/setup-buildx-action#435

Full Changelog: docker/setup-buildx-action@v3.11.1...v3.12.0

v3.11.1

• Fix keep-state not being respected by @​crazy-max in docker/setup-buildx-action#429

Full Changelog: docker/setup-buildx-action@v3.11.0...v3.11.1

v3.11.0

• Keep BuildKit state support by @​crazy-max in docker/setup-buildx-action#427
• Remove aliases created when installing by default by @​hashhar in docker/setup-buildx-action#139
• Bump @​docker/actions-toolkit from 0.56.0 to 0.62.1 in docker/setup-buildx-action#422 docker/setup-buildx-action#425

Full Changelog: docker/setup-buildx-action@v3.10.0...v3.11.0

v3.10.0

• Bump @​docker/actions-toolkit from 0.54.0 to 0.56.0 in docker/setup-buildx-action#408

Full Changelog: docker/setup-buildx-action@v3.9.0...v3.10.0

v3.9.0

• Bump @​docker/actions-toolkit from 0.48.0 to 0.54.0 in docker/setup-buildx-action#402 docker/setup-buildx-action#404

Full Changelog: docker/setup-buildx-action@v3.8.0...v3.9.0

v3.8.0

• Make cloud prefix optional to download buildx if driver is cloud by @​crazy-max in docker/setup-buildx-action#390
• Bump @​actions/core from 1.10.1 to 1.11.1 in docker/setup-buildx-action#370
• Bump @​docker/actions-toolkit from 0.39.0 to 0.48.0 in docker/setup-buildx-action#389
• Bump cross-spawn from 7.0.3 to 7.0.6 in docker/setup-buildx-action#382

Full Changelog: docker/setup-buildx-action@v3.7.1...v3.8.0

... (truncated)

Commits

• d7f5e7f Merge pull request #489 from docker/dependabot/npm_and_yarn/docker/actions-to...
• 92bc5c9 chore: update generated content
• da11e35 build(deps): bump @​docker/actions-toolkit from 0.79.0 to 0.90.0
• f021e16 Merge pull request #492 from docker/dependabot/npm_and_yarn/undici-6.24.1
• b5af94f chore: update generated content
• 16ad977 build(deps): bump undici from 6.23.0 to 6.25.0
• d7a12d7 Merge pull request #495 from docker/dependabot/npm_and_yarn/glob-10.5.0
• 28ff27d build(deps): bump glob from 10.3.12 to 13.0.6
• daf436b Merge pull request #496 from docker/dependabot/npm_and_yarn/fast-xml-parser-5...
• 9725348 chore: update generated content
• Additional commits viewable in compare view

[github-actions]

NeuralMind self-benchmark

Status: PASS — floor 4×, measured 6.6×.

Phase 1 — Reduction on committed fixture

• Average reduction: 6.6×
• Top-k retrieval hit rate: 71.7%
• Naive baseline: 47,360 tokens (all fixture files concatenated)
• NeuralMind total: 7,296 tokens across 10 queries
• Estimated monthly savings @ 100 queries/day on Claude 3.5 Sonnet: ~$36.06

#	Query	Shape	Naive	NeuralMind	Ratio	Hit
1	auth-flow	cross-file	4,736	726	6.5×	33.3%
2	api-endpoints	focused	4,736	714	6.6×	100.0%
3	billing-flow	cross-file	4,736	710	6.7×	33.3%
4	user-storage	cross-file	4,736	825	5.7×	50.0%
5	jwt-verify	focused	4,736	590	8.0×	100.0%
6	stripe-webhook	focused	4,736	599	7.9×	100.0%
7	create-user	cross-file	4,736	855	5.5×	50.0%
8	refund	focused	4,736	732	6.5×	100.0%
9	db-choice	identity	4,736	834	5.7×	100.0%
10	invoice-send	cross-file	4,736	711	6.7×	50.0%

Phase 2 — Learning uplift

• Memory events logged: 20
• Learned patterns: 14
• Reduction ratio after neuralmind learn: 6.5× (Δ -0.09× vs. cold)
• Top-k hit rate after learning: 75.0% (Δ +3.3 points vs. cold)

Note: uplift numbers on a 500-line fixture are intentionally modest — the point is to
verify the learning mechanism persists and applies. On real production repos the lift
is larger; this test only catches regressions in persistence.

Phase 3 — Synapse recall A/B (same warm graph, recall off vs on)

• Synapse edges after seeding co-editing sessions: 3610
• Top-k hit rate: 71.7% off → 83.3% on (Δ +11.7 points)
• Reduction ratio: 6.6× off → 6.6× on (Δ -0.07× — budget-neutral by design)

This isolates the Hebbian synapse layer from the learned_patterns reranker in
Phase 2. The hit-rate delta shows associative recall surfacing co-edited modules a
purely textual search ranks lower; the near-zero reduction delta confirms it does so
without spending extra tokens (recalled nodes displace the weakest hits, not add to them).

Assumptions

• Baseline: every .py file in tests/fixtures/sample_project/ concatenated.
• Tokenizer: tiktoken GPT-4o encoding (per-model breakdown in multi_model.json if generated).
• Pricing: Claude 3.5 Sonnet input @ $3.0/MTok.
• Regression floor: 4× — well below NeuralMind's typical 40–70× on real repos.

Per-model token reduction

Model	Tokenizer	Naive	NeuralMind	Ratio	Source
GPT-4o / GPT-4o-mini	tiktoken o200k_base	4,739	779	6.1×	measured
GPT-4 / GPT-3.5-turbo	tiktoken cl100k_base	4,710	770	6.1×	measured
Claude 3.5 Sonnet	estimated: GPT-4o × 1.08 — install anthropic for an exact count	5,118	841	6.1×	estimated
Llama 3 (70B)	estimated: GPT-4o × 1.22 — Llama tokenizer requires model weights; estimate based on published vocab ratios	5,781	950	6.1×	estimated

Rows marked measured use the provider's real tokenizer. Rows marked
estimated apply a published vocab-size correction to the GPT-4o count —
honest approximations, not hardcoded claims.

---

Automated by .github/workflows/ci-benchmark.yml — regenerate locally with python -m tests.benchmark.run.

[dfrostar]

@dependabot rebase

---

Generated by Claude Code