Back to leaderboard

qwen3.5-9b

LM-STUDIO GGUF

qwen35 · 9B · Q4_K_S

Not Rec.

Mar 14, 2026 · Apple M4

qwen3.8-27b-mlx

LM-STUDIO MLX

qwen3_5 · 27B · 4bit

Good

Oct 20, 2026 · Apple M1 Pro

Global Score
72 vs 78
Hardware Fit
51 vs 56
Quality Score
81 vs 87

Hardware

qwen3.5-9b qwen3.8-27b-mlx
MachineMacBook AirMacBook Pro
CPUApple M4Apple M1 Pro
Cores1010
RAM32 GB32 GB
GPUApple M4Apple M1 Pro
OSmacOS 26.3macOS 15.6.1
Archarm64arm64
Power Modebalancedbalanced

Performance

qwen3.5-9b qwen3.8-27b-mlx
Tokens/sec14.410.8
First chunk4 ms8 ms
TTFT824 ms2.6 s
Load timeN/AN/A
Memory usage48.3 GB21.0 GB
Memory %151%66%

HW Fit Score Breakdown

qwen3.5-9b

Speed
31/50
TTFT
20/20
Memory
0/30

qwen3.8-27b-mlx

Speed
23/50
TTFT
16/20
Memory
17/30

Quality

qwen3.5-9b

Reasoning
16/20
Coding
15/20
Instruction
16/20
Structured
15/15
Math
9/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Adequate Multilingual: Strong

qwen3.8-27b-mlx

Reasoning
18/20
Coding
18/20
Instruction
15/20
Structured
15/15
Math
11/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Adequate Multilingual: Strong
View qwen3.5-9b details | View qwen3.8-27b-mlx 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