Back to leaderboard

qwen3.8-27b-mlx

LM-STUDIO MLX

qwen3_5 · 27B · 8bit

Marginal

Aug 17, 2026 · Apple M5 Max

qwen3.5:35b-a3b-q4_K_M

OLLAMA GGUF

qwen35moe · 36.0B · Q4_K_M

Excellent

Aug 22, 2026 · Intel Xeon® E5-2637 v3

Global Score
59 vs 90
Hardware Fit
63 vs 98
Quality Score
58 vs 86

Hardware

qwen3.8-27b-mlx qwen3.5:35b-a3b-q4…
MachineMacBook ProDell Inc. Precision Tower 7910
CPUApple M5 MaxIntel Xeon® E5-2637 v3
Cores1816
RAM64 GB96 GB
GPUApple M5 MaxNVIDIA GeForce RTX 3090, NVIDIA GeForce RTX 3060
OSmacOS 27.0Microsoft Windows 10 Pro 10.0.19045
Archarm64x64
Power Modebalancedbalanced

Performance

qwen3.8-27b-mlx qwen3.5:35b-a3b-q4…
Tokens/sec18.348.4
First chunk4 ms1085 ms
TTFT530 ms1.1 s
Load timeN/A22.2 s
Memory usage38.5 GB22.9 GB
Memory %60%24%

HW Fit Score Breakdown

qwen3.8-27b-mlx

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

qwen3.5:35b-a3b-q4_K_M

Speed
50/50
TTFT
18/20
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

qwen3.5:35b-a3b-q4_K_M

Reasoning
16/20
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: Strong Multilingual: Strong
View qwen3.8-27b-mlx details | View qwen3.5:35b-a3b-q4… 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