Back to leaderboard

qwen3.8-27b-mlx

LM-STUDIO MLX

qwen3_5 · 27B · 8bit

Marginal

Aug 17, 2026 · Apple M5 Max

GPT-120B:120b

OLLAMA GGUF

gptoss · 116.8B · MXFP4

Not Rec.

Aug 22, 2026 · NVIDIA RTX Spark N1X (6144-core GPU, 20-core ) RTX Spark N1X (6144-core GPU, 20-core )

Global Score
59 vs 66
Hardware Fit
63 vs 23
Quality Score
58 vs 84

Hardware

qwen3.8-27b-mlx GPT-120B:120b
MachineMacBook ProHP HyperX OMEN Transcend Gaming Laptop 14-ga0xxx5CD612DXML
CPUApple M5 MaxNVIDIA RTX Spark N1X (6144-core GPU, 20-core ) RTX Spark N1X (6144-core GPU, 20-core )
Cores1820
RAM64 GB55 GB
GPUApple M5 MaxNVIDIA RTX Spark N1X (6144-core Blackwell RTX GPU), NVIDIA NPU
OSmacOS 27.0Microsoft Windows 11 Pro 10.0.28000
Archarm64arm64
Power Modebalancedbalanced

Performance

qwen3.8-27b-mlx GPT-120B:120b
Tokens/sec18.316.8
First chunk4 ms11247 ms
TTFT530 ms12.2 s
Load timeN/A77.3 s
Memory usage38.5 GB61.6 GB
Memory %60%113%

HW Fit Score Breakdown

qwen3.8-27b-mlx

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

GPT-120B:120b

Speed
21/50
TTFT
2/20
Memory
0/30

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

GPT-120B:120b

Reasoning
16/20
Coding
19/20
Instruction
11/20
Structured
13/15
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