🤫husshhussh
🤫husshhusshOnePuppy
🤫Puppy OneReserve
← Back to the series

🤫 Knowledge, Research & Investigation · Note 001

The smallest machine that completes the job wins.

A pre-ship thesis on Apple M6, M5 Pro, M5 Max, and M5 Ultra for private agentic AI - and the measurement system hussh will use to find the best useful work per watt and per dollar.

August 27, 2026Pre-ship analysis18 minute read

Evidence status. The new Macs ship September 22; the 512GB Studio follows in late October. M6 and M5 Ultra do not yet have independent output-token or wall-power measurements. Every forward-looking number below is marked as an estimate.

We spent the first era of AI asking what a model could do.

The next era begins with a more human question: where should intelligence live?

Cloud AI has made extraordinary capability available to almost anyone. But distance has a cost. A request travels. A token becomes a recurring charge. Personal context may leave the device where it began.

For a personal agent - one that helps with the small, meaningful work of everyday life - we believe the default should be closer. Close to you. Close to your data. Close to the moment when help is needed.

That is the idea behind the new 🤫 Puppy One series: deliberate Apple-silicon configurations, guided Agent One activation, transparent permissions, and a routing principle that begins on device and goes farther only when the user approves and the job truly benefits.

Your device first.
Your permission always.

The answer in 60 seconds

Our recommended launch line.

01

Puppy One 32

M6 · 32GB

Personal efficiency candidate
02

Puppy One 64

M5 Pro · 64GB

Choose for the mini footprint
03

Puppy One 128

M5 Max · 128GB

Best professional balance
04

Puppy One 256

M5 Ultra · 256GB

Large-model workstation
05

Puppy One 512

M5 Ultra · 512GB

Capacity flagship, only when needed

The commercial sweet spot: Mac Studio with the 40-core M5 Max, 64GB or 128GB of unified memory, and 1TB or 2TB storage.

The efficiency candidate: Mac mini with M6 and 32GB. We expect it to lead personal, always-on workloads, but we will not call it the winner until wall-power testing is complete.

I · What changed

A five-inch efficiency story. A 512GB capacity story.

Apple's August releases widen the local-AI design space at both ends. M6 combines a 2-nanometer process, a 12-core CPU, a 12-core GPU with Neural Accelerators, dual 16-core Neural Engines, and up to 170GB/s in Mac mini. M5 Ultra combines four dies through UltraFusion, an up-to-80-core GPU, a 32-core Neural Engine, 1.2TB/s of unified-memory bandwidth, and as much as 512GB of memory in Mac Studio. Apple silicon release ↗

hussh tierApple configurationCPU / GPUMemoryBandwidthPrimary job
Puppy One 32Mac mini · M612 / 12 cores32GB170GB/s4B-14B specialists; personal router
Puppy One 64Mac mini · M5 Pro18 / 20 cores64GB307GB/s27B-32B; compact professional
Puppy One 128Mac Studio · M5 Max18 / 40 cores128GB614GB/s70B; context and concurrency
Puppy One 256Mac Studio · M5 Ultra36 / 80 cores256GB1.2TB/s120B-235B; research and teams
Puppy One 512Mac Studio · M5 Ultra36 / 80 cores512GB1.2TB/s300B+ quantized experiments

Model classes are planning ranges, not guarantees. Fit depends on architecture, weight precision, KV-cache precision, context length, runtime overhead, and concurrency. Recommended M6 32GB configurations receive 170GB/s; Apple lists 153GB/s on standard configurations. Mac mini specs ↗ · Mac Studio specs ↗

II · Hardware is only half the product

Apple is turning the operating system into an agent surface.

Apple's software direction is as consequential as the silicon. The opportunity is not "Siri inside hussh." It is a set of supported, permissioned interfaces through which Agent One can run local models, expose its own actions, and cooperate with system experiences without pretending to be part of Apple Intelligence.

Available now

Stable foundations on macOS 26

  • Foundation Models access to Apple's OS-managed on-device language model.
  • App Intents for hussh-declared actions and entities in Shortcuts and Spotlight.
  • Core ML for optimized packaged models.
  • MLX and MLX-LM for open-weight local inference and fine-tuning on Apple silicon.
Beta · macOS 27

The next agentic layer

  • AFM 3 Core and the sparse, multimodal 20B AFM 3 Core Advanced.
  • Expanded Foundation Models APIs for images, dynamic profiles, custom providers, and eligible Private Cloud Compute access.
  • Core AI for bringing custom full-scale LLMs on device.
  • New Siri AI, App Intents schemas, personal context, onscreen awareness, and app actions.

AFM 3 Core Advanced is especially interesting technically: Apple describes a 20-billion-parameter sparse model that activates 1-4 billion parameters per request, storing the full model in flash and selecting experts per prompt to reduce DRAM pressure. That is a different optimization path from simply buying more memory for a dense open model. Apple Foundation Models research ↗

What App Intents does not do.

It does not grant Agent One arbitrary access to Mail, Photos, Messages, Siri history, or other apps. It makes only declared actions and entities available through supported schemas. Any network tool call also changes the privacy boundary. "On device" is a property of a tested execution path, not a blanket label for the whole agent.

As of August 27, macOS 27, Core AI, expanded Foundation Models features, and Siri AI remain pre-release. Apple says macOS 27 ships this fall, with feature, language, region, entitlement, and device limits. Apple developer release ↗ · Apple Intelligence for developers ↗

III · The metric

Token per watt is a unit.
Useful work is the product.

Token efficiency
tokens / second÷joules / second=tokens / joule

The inverse is joules per token. A "watt per token" dashboard is usually describing the same energy cost with imprecise units.

Decode throughput matters. But an agent also ingests prompts, embeds documents, retrieves context, invokes tools, waits on networks, checks results, and sometimes retries. A small model that produces tokens rapidly but fails the task can look efficient while wasting the user's energy.

Our north star is therefore successful tasks per joule, paired with successful tasks per dollar. Raw prompt throughput, output throughput, and token energy remain the auditable systems layer beneath it.

SystemsPrompt tok/s · Decode tok/s · Token/J · J/token
ExperienceTTFT · p50/p95 latency · thermal stability
AgentTool success · accepted answers · retry rate
BusinessTasks/kWh · Tasks/$ · local completion rate

IV · What can be modeled before shipment

Decode is usually a memory problem. Prefill is usually a compute problem.

For a dense, quantized model generating one token at a time, the GPU repeatedly reads most model weights. A useful first-order roofline is:

TPS ≈ ηBW × memory bandwidth ÷ bytes touched per generated token

For dense Q4 weights at short context, a planning assumption is roughly 0.55GB per billion parameters, plus runtime and KV cache. Observed Apple-silicon bandwidth utilization varies by chip, model, and runtime; 55-80% is a useful range, not a promise. Longer context increases KV traffic and can dominate.

That context penalty is not theoretical. A systematic MLX M3 Ultra benchmark saw Qwen 32B Q4 decline from 31.2 tok/s at 1K context to 8.5 tok/s at 128K. The right page cannot show one heroic token number; it must show a curve. MLX context study ↗

Pre-ship batch-one decode model

llama.cpp / Metal · warm run · short context · output tokens per second. Estimates are intentionally broad. They are not PuppyBench results.

Dense Q4 modelM6 · 170GB/sM5 Pro · 307GB/sM5 Max · 614GB/sM5 Ultra · 1.2TB/s
7B27-36est.53-6563.86 community run115-125119.92 measured170-235est.
27B6-9est.11-16est.23-32est.44-62est.
70BDoes not fit safely4-6est.8-12est.17-23est.
120BDoes not fitDoes not fit safely5-7est.10-14est.

M5 Max reference: maintained llama.cpp Apple-silicon table, Llama 2 7B Q4, 40-core GPU / 128GB. M5 Pro reference is a community result on an older build and is not strictly comparable. M6 and M5 Ultra rows remain empty before hardware availability. llama.cpp measurements ↗

V · Best bang for the buck

Pay for the memory the job needs - then buy bandwidth.

Capacity is a gate. If the model and its context do not fit, the machine is not merely slower; it is the wrong machine. Once the workload fits, bandwidth and runtime determine decode speed, while Neural Accelerators and GPU compute matter heavily for prompt processing and other matrix-rich work.

Most people

M6 · 32GB / 1TB

Our personal, always-on candidate. Best for smaller specialist models, retrieval, routing, voice, and one active user. Treat 27B as an edge case, not the default.

Small footprint

M5 Pro · 64GB / 1TB

Capable 32B machine, but economically squeezed: the observed Apple price is only about $600 below a 40-core M5 Max / 64GB Studio with roughly twice the bandwidth. Buy Pro when the five-inch form factor matters.

Professional sweet spot

M5 Max · 40 GPU / 64GB or 128GB

The strongest balance for coding, RAG, 27B-70B models, context headroom, and 1-4 concurrent flows. Move to 128GB when the model or context actually requires it.

Absolute throughput

M5 Ultra · 64 GPU / 96GB

Both Ultra GPU tiers list 1.2TB/s. The base Ultra may become the best single-stream decode-bandwidth per dollar, while supporting several smaller agents. It needs post-ship verification.

Memory first

M5 Ultra · 80 GPU / 256GB or 512GB

For 120B-235B models, large KV caches, high concurrency, or compute-heavy prefill. Choose 512GB for 300B+ quantized experiments or fine-tuning - not ordinary personal-agent value.

Observed Apple U.S. configuration prices · August 27, 2026

M6 32GB/1TB: $1,699 · M5 Pro 18/20 64GB/1TB: $3,199 · M5 Max 40 GPU 64GB/1TB: $3,799 · M5 Max 40 GPU 128GB/1TB: $5,399 · M5 Ultra 80 GPU 256GB/2TB: $11,299.

Verify live before publication or sale. These are Apple hardware prices observed during research, not hussh pricing, and exclude taxes, support, setup, software, and services.

VI · Token economics

Utilization and capital cost dominate electricity.

Capex per 1M tokenspurchase price ÷ (TPS × lifetime seconds × utilization ÷ 1M)
Energy per 1M tokensJ/token × 0.27778 × electricity $/kWh

At $0.15/kWh, even 0.35-1.0 joule per token costs only about $0.015-$0.042 per million output tokens. Electricity matters for thermals, sustainability, and always-on fleets, but for a personal device, capital utilization is usually the larger economic lever.

A five-year machine at 35% inference utilization produces a capex denominator of 55.188 × TPS. That is why a memory upgrade bought "just in case" can be expensive: it may add capacity without adding single-stream bandwidth or useful throughput. The right configuration is the one you can keep productively occupied.

VII · PuppyBench

No mystery score.
No benchmark theatre.

Once shipping hardware arrives, hussh should publish a reproducible qualification book for each recommended configuration. Every result should carry the exact model, hash, precision, context, software, power method, and uncertainty.

Models

4B / 8B / 14B / 27B / 70B dense Q4; 120B MoE; Q8 or BF16 where fit.

Context

1K / 8K / 32K / 128K. Prompts at 512 / 4K / 32K. Generate 256 and 1,024 tokens.

Concurrency

1 / 2 / 4 / 8 streams. Report aggregate and per-user throughput.

Latency

Cold and warm TTFT; prefill; decode; p50/p95 inter-token latency.

Power

Calibrated AC wall meter plus system telemetry. Report gross and idle-subtracted energy.

Rigor

Ten repeats, median, p95, 95% confidence interval, plus a 30-minute sustained thermal run.

Quality

Fixed agent suite: retrieval, tool use, App Intent completion, answer acceptance, retries.

Cost

Hardware, support, electricity, cloud fallback, and successful tasks per dollar.

Publish the method beside the number. A benchmark customers cannot reproduce is an advertisement, not evidence.

VIII · The partnership boundary

Built for Apple silicon.
Independently hussh.

The public record supports a strong, truthful story: hussh can curate Apple hardware, build Agent One with public Apple developer frameworks, and serve customers who choose Mac. The three Apple releases do not name hussh or establish an Apple design or launch partnership.

Until an executed agreement supplies approved external wording, the product should say "configured by hussh on Apple silicon", not "co-developed with Apple," "mutual customers," or "Apple launch partner." That distinction protects customer trust and makes the real product achievement more credible.

It also protects the design. The goal is not to copy Apple's website or imply that anyone at Apple approved it. The goal is to apply the deeper discipline: one clear idea, ruthless editing, honest materials, and no claim that the product cannot carry.

Apple's third-party trademark guidelines ↗

The decision

Your agent deserves a home.

Choose Puppy One 32 for efficient everyday assistance.

Choose Puppy One 64 when the five-inch footprint and professional headroom matter.

Choose Puppy One 128 when the Mac becomes a serious local AI studio.

Choose Puppy One 256 or 512 only when model capacity, context, and data locality justify the additional capital.

The best personal supercomputer is not the one with the largest number. It is the one that quietly gets the right job done.

Your data. Your agent. Your business.
Compare the seriesReserve from $0.01

Primary sources & technical evidence

Apple M6 and M5 Ultra ↗Apple Mac mini release ↗Apple Mac Studio release ↗Mac mini specifications ↗Mac Studio specifications ↗Third-generation Apple Foundation Models ↗Apple developer frameworks ↗Apple MLX on M5 ↗llama.cpp Apple benchmarks ↗Apple-silicon power methodology ↗MLX context-length study ↗

Independent pre-ship analysis by hussh. Apple hardware claims and dates are attributed to Apple. hussh model-fit and performance ranges are estimates unless identified as measured. Apple, Mac, Mac mini, Mac Studio, macOS, Siri, Apple Intelligence, and related marks are trademarks of Apple Inc. hussh is independent; this report is not produced, sponsored, or endorsed by Apple.