Back to leaderboard

deepseek-r1:1.5b

OLLAMA GGUF

qwen2 · 1.8B · Q4_K_M

Good

Mar 6, 2026 · Apple M4

qwen3.5-9b-mlx

LM-STUDIO MLX

qwen3_5 · 9B · 4bit

Good

Mar 12, 2026 · Apple M4 Pro

Global Score
66 vs 77
Hardware Fit
100 vs 67
Quality Score
51 vs 81

Hardware

deepseek-r1:1.5b qwen3.5-9b-mlx
MachineMacBook AirMac mini
CPUApple M4Apple M4 Pro
Cores1014
RAM32 GB64 GB
GPUApple M4Apple M4 Pro
OSmacOS 26.3macOS 15.7.4
Archarm64arm64
Power Modebalancedbalanced

Performance

deepseek-r1:1.5b qwen3.5-9b-mlx
Tokens/sec85.420.3
First chunk115 ms4216 ms
TTFT115 ms4.2 s
Load time0.5 sN/A
Memory usage2.6 GB5.6 GB
Memory %8%9%

HW Fit Score Breakdown

deepseek-r1:1.5b

Speed
50/50
TTFT
20/20
Memory
30/30

qwen3.5-9b-mlx

Speed
29/50
TTFT
8/20
Memory
30/30

Quality

deepseek-r1:1.5b

Reasoning
14/20
Coding
4/20
Instruction
8/20
Structured
6/15
Math
14/15
Multilingual
5/10
Reasoning: Adequate Coding: Poor Instruction Following: Weak Structured Output: Weak Math: Strong Multilingual: Adequate

qwen3.5-9b-mlx

Reasoning
15/20
Coding
18/20
Instruction
16/20
Structured
14/15
Math
8/15
Multilingual
10/10
Reasoning: Adequate Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Adequate Multilingual: Strong

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