Run DeepSeek R1 Distill 8B on Apple Silicon

DeepSeek R1 Distill 8B brings o1-style chain-of-thought reasoning to a model that runs on the cheapest Apple Silicon config: we measured 20.0 tokens/sec on a base M4 with 16 GB (Q4_K_M, median of 3 runs). Budget extra tokens per answer — reasoning traces are long — but for math, logic, and stepwise analysis it punches far above its size.

Parameters
8B (distill)
Min memory (Q4_K_M)
16 GB
From
$99/mo

How much memory does R1 8B need?

Weight size is parameter count × bits per weight; recommended memory adds 15% runtime headroom plus a KV-cache allowance for the 131,072-token context window.

QuantizationWeightsRecommended RAMCheapest fitNotes
Q4_K_M~4.9 GB16 GBM4 (Mac mini)Best quality/size trade-off — the default for most deployments
Q8_0~8.5 GB16 GBM4 (Mac mini)Near-lossless; noticeably better on code and long reasoning
FP16~16 GB24 GBM4 (Mac mini)Full precision — research and eval baselines

Which Mac should you pick?

Generation speed on Apple Silicon scales mostly with memory bandwidth, so the chip choice sets both whether the model fits and how fast it runs.

ChipMax RAMBandwidthFits R1 8B?From
M4 · Mac mini32 GB120 GB/s Q4 + Q8 + FP16$99/mo
M4 Pro · Mac mini64 GB273 GB/s Q4 + Q8 + FP16$199/mo
M4 Max · Mac Studio128 GB546 GB/s Q4 + Q8 + FP16$286/mo
M3 Ultra · Mac Studio256 GB819 GB/s Q4 + Q8 + FP16$572/mo

Deploy R1 8B in about 5 minutes

  1. 1

    Build a Mac in the constructor — pick an M4 with at least 16 GB of unified memory.

  2. 2

    Pick any Ollama-based stack (or a clean machine) and pull DeepSeek R1 Distill 8B with one command — every deployment ships the OpenAI-compatible API.

  3. 3

    Point your existing OpenAI SDK code at the new base_url and ship.

from openai import OpenAI

client = OpenAI(
    api_key="mcy_live_...",
    base_url="https://dep-xxxx.macyou.cloud/v1",
)
response = client.chat.completions.create(
    model="deepseek-r1-8b",
    messages=[{"role": "user", "content": "Hello!"}],
)

License and what it's good at

MIT (distill weights); base model licenses apply The R1 distill weights are MIT-licensed; the Llama base carries Meta's community license terms. Made by DeepSeek.

Typical workloads: math and logic-heavy assistants; code review with visible reasoning; evaluating whether reasoning models help your workload before paying for 70b.

Frequently asked questions

Why is R1 8B slower per answer than Llama 8B at similar tok/s?

Generation speed is nearly identical (20.0 vs 21.2 tok/s measured), but R1 emits a chain of thought before the answer — often 2–5× more tokens per response. You pay in latency, not in speed.

Is there a bigger single-machine option?

Yes — the R1 Distill 70B (Llama-3.3 base) at Q4 fits a 64 GB M4 Pro. See our post “DeepSeek V3 70B doesn't exist — run the R1 Distill 70B instead” for that setup.

Related