phi-3.5-mini-instruct

LM-STUDIO MLX

phi3 · 4bit

Marginal

Mar 4, 2026 · Apple M4

deepseek-r1:70b

OLLAMA GGUF

llama · 70.6B · Q4_K_M

Not Rec.

Jun 27, 2026 · Cortex-X925

Global Score

55 vs 47

Hardware Fit

100 vs 51

Quality Score

36 vs 45

Hardware

phi-3.5-mini-instr… deepseek-r1:70b

MachineMacBook AirNVIDIA NVIDIA_DGX_Spark

CPUApple M4Cortex-X925

Cores1020

RAM32 GB122 GB

GPUApple M4Device 2e12

OSmacOS 26.3Ubuntu 24.04.4 LTS

Archarm64arm64

Power Modebalancedperformance

Performance

phi-3.5-mini-instr… deepseek-r1:70b

Tokens/sec43.44.7

First chunk359 ms557 ms

TTFT359 ms557 ms

Load timeN/A52.6 s

Memory usage2.0 GB49.5 GB

Memory %6%41%

HW Fit Score Breakdown

phi-3.5-mini-instruct

Speed

50/50

TTFT

20/20

Memory

30/30

deepseek-r1:70b

Speed

4/50

TTFT

20/20

Memory

27/30

Quality

phi-3.5-mini-instruct

Reasoning

11/20

Coding

4/20

Instruction

7/20

Structured

1/15

Math

7/15

Multilingual

6/10

Reasoning: Adequate Coding: Poor Instruction Following: Weak Structured Output: Poor Math: Weak Multilingual: Adequate

deepseek-r1:70b

Reasoning

3/20

Coding

1/20

Instruction

14/20

Structured

9/15

Math

8/15

Multilingual

10/10

Reasoning: Poor Coding: Poor Instruction Following: Adequate Structured Output: Adequate Math: Adequate Multilingual: Strong

View phi-3.5-mini-instr… details | View deepseek-r1:70b details

Run yours and compare

$ npm install -g metrillm@latest

$ metrillm

Requires Node 20+ and Ollama or LM Studio running

Or run without installing: npx metrillm@latest