Back to leaderboard

qwen/qwen3.8-27b

LM-STUDIO MLX

qwen3_5 · 27B · 8bit

Excellent

Aug 18, 2026 · Apple M4 Max

granite4.2:3b

OLLAMA GGUF

granite · 3.7B · Q4_K_M

Excellent

Aug 29, 2026 · Apple M1 Pro

Global Score
81 vs 80
Hardware Fit
68 vs 80
Quality Score
87 vs 80

Hardware

qwen/qwen3.8-27b granite4.2:3b
MachineMacBook ProMacBook Pro
CPUApple M4 MaxApple M1 Pro
Cores1610
RAM128 GB16 GB
GPUApple M4 MaxApple M1 Pro
OSmacOS 26.5.2macOS 26.6.2
Archarm64arm64
Power Modebalancedbalanced

Performance

qwen/qwen3.8-27b granite4.2:3b
Tokens/sec16.446.8
First chunk5 ms135 ms
TTFT742 ms135 ms
Load timeN/A0.0 s
Memory usage38.5 GB12.7 GB
Memory %30%79%

HW Fit Score Breakdown

qwen/qwen3.8-27b

Speed
18/50
TTFT
20/20
Memory
30/30

granite4.2:3b

Speed
50/50
TTFT
20/20
Memory
10/30

Quality

qwen/qwen3.8-27b

Reasoning
17/20
Coding
18/20
Instruction
16/20
Structured
15/15
Math
11/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Adequate Multilingual: Strong

granite4.2:3b

Reasoning
18/20
Coding
13/20
Instruction
12/20
Structured
15/15
Math
14/15
Multilingual
8/10
Reasoning: Strong Coding: Adequate Instruction Following: Adequate Structured Output: Strong Math: Strong 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