Back to leaderboard

qwen3.8-27b-mlx

LM-STUDIO MLX

qwen3_5 · 27B · 8bit

Marginal

Aug 17, 2026 · Apple M5 Max

gptoss · 20.9B · MXFP4

OLLAMA GGUF

gptoss · 20.9 · MXFP4

Excellent

Jul 9, 2026 · Intel Core™ i5-10400

Global Score
59 vs 84
Hardware Fit
63 vs 69
Quality Score
58 vs 91

Hardware

qwen3.8-27b-mlx gptoss · 20.9B · M…
MachineMacBook ProMac Pro
CPUApple M5 MaxIntel Core™ i5-10400
Cores1812
RAM64 GB32 GB
GPUApple M5 MaxAMD Radeon RX 6700 XT
OSmacOS 27.0macOS 26.5.1
Archarm64x64
Power Modebalancedbalanced

Performance

qwen3.8-27b-mlx gptoss · 20.9B · M…
Tokens/sec18.317.5
First chunk4 ms643 ms
TTFT530 ms7.7 s
Load timeN/A0.0 s
Memory usage38.5 GB0.8 GB
Memory %60%3%

HW Fit Score Breakdown

qwen3.8-27b-mlx

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

gptoss · 20.9B · MXFP4

Speed
34/50
TTFT
5/20
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

gptoss · 20.9B · MXFP4

Reasoning
19/20
Coding
19/20
Instruction
13/20
Structured
15/15
Math
15/15
Multilingual
10/10
Reasoning: Strong Coding: Strong 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