Back to leaderboard

qwen3-30b-a3b-thinking-2507-claude-4.5-sonnet-high-reasoning-distill-mlx

LM-STUDIO MLX

qwen3_moe · 30B · MXFP4

Excellent

Apr 20, 2026 · Apple M2 Max

qwen3.5-2b-mlx

LM-STUDIO MLX

qwen3_5 · 2B · 4bit

Good

Mar 6, 2026 · Apple M4 Pro

Global Score
97 vs 65
Hardware Fit
95 vs 88
Quality Score
98 vs 55

Hardware

qwen3-30b-a3b-thin… qwen3.5-2b-mlx
MachineMac StudioMac mini
CPUApple M2 MaxApple M4 Pro
Cores1214
RAM96 GB64 GB
GPUApple M2 MaxApple M4 Pro
OSmacOS 26.1macOS 15.7.4
Archarm64arm64
Power Modebalancedbalanced

Performance

qwen3-30b-a3b-thin… qwen3.5-2b-mlx
Tokens/sec79.648.9
First chunk11 ms4208 ms
TTFT1.7 s4.2 s
Load time7.1 sN/A
Memory usage0.0 GB1.6 GB
Memory %0%3%

HW Fit Score Breakdown

qwen3-30b-a3b-thinking-…

Speed
50/50
TTFT
15/20
Memory
30/30

qwen3.5-2b-mlx

Speed
50/50
TTFT
8/20
Memory
30/30

Quality

qwen3-30b-a3b-thinking-…

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

qwen3.5-2b-mlx

Reasoning
10/20
Coding
7/20
Instruction
13/20
Structured
14/15
Math
3/15
Multilingual
8/10
Reasoning: Adequate Coding: Weak Instruction Following: Adequate Structured Output: Strong Math: Poor 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