Back to leaderboard

qwen3.8-27b-mlx

LM-STUDIO MLX

qwen3_5 · 27B · 8bit

Marginal

Aug 17, 2026 · Apple M5 Max

granite3.1-moe:3b

OLLAMA GGUF

granitemoe · 3.3B · Q4_K_M

Good

May 24, 2026 · Intel Gen Intel® Core™ i7-1355U

Global Score
59 vs 66
Hardware Fit
63 vs 99
Quality Score
58 vs 52

Hardware

qwen3.8-27b-mlx granite3.1-moe:3b
MachineMacBook ProDell Inc. Inspiron 15 3530
CPUApple M5 MaxIntel Gen Intel® Core™ i7-1355U
Cores1812
RAM64 GB31 GB
GPUApple M5 MaxRaptor Lake-P [Iris Xe Graphics]
OSmacOS 27.0cachyos rolling
Archarm64x64
Power Modebalancedlow-power

Performance

qwen3.8-27b-mlx granite3.1-moe:3b
Tokens/sec18.327.2
First chunk4 ms527 ms
TTFT530 ms527 ms
Load timeN/A1.9 s
Memory usage38.5 GB2.9 GB
Memory %60%9%

HW Fit Score Breakdown

qwen3.8-27b-mlx

Speed
23/50
TTFT
20/20
Memory
20/30

granite3.1-moe:3b

Speed
49/50
TTFT
20/20
Memory
30/30

Quality

qwen3.8-27b-mlx

Reasoning
19/20
Coding
14/20
Instruction
6/20
Structured
1/15
Math
13/15
Multilingual
5/10
Reasoning: Strong Coding: Adequate Instruction Following: Weak Structured Output: Poor Math: Strong Multilingual: Weak

granite3.1-moe:3b

Reasoning
7/20
Coding
9/20
Instruction
13/20
Structured
12/15
Math
5/15
Multilingual
6/10
Reasoning: Weak Coding: Weak Instruction Following: Adequate Structured Output: Strong Math: Weak Multilingual: Adequate

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