Back to leaderboard

deepseek-r1-distill-qwen-14b-mlx

LM-STUDIO MLX

qwen2 · 14B · 5bit

Not Rec.

Mar 11, 2026 · Apple M4

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

LM-STUDIO MLX

qwen3_moe · 30B · MXFP4

Excellent

Aug 20, 2026 · Apple M4 Max

Global Score
36 vs 96
Hardware Fit
53 vs 95
Quality Score
28 vs 96

Hardware

deepseek-r1-distil… qwen3-30b-a3b-thin…
MachineMac miniMacBook Pro
CPUApple M4Apple M4 Max
Cores1016
RAM24 GB128 GB
GPUApple M4Apple M4 Max
OSmacOS 26.3macOS 26.5.2
Archarm64arm64
Power Modebalancedbalanced

Performance

deepseek-r1-distil… qwen3-30b-a3b-thin…
Tokens/sec10.3112.8
First chunk27 ms21 ms
TTFT20.7 s1.5 s
Load time15.4 s3.5 s
Memory usage1.0 GB0.0 GB
Memory %4%0%

HW Fit Score Breakdown

deepseek-r1-distill-qwe…

Speed
23/50
TTFT
0/20
Memory
30/30

qwen3-30b-a3b-thinking-…

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

Quality

deepseek-r1-distill-qwe…

Reasoning
3/20
Coding
6/20
Instruction
6/20
Structured
3/15
Math
2/15
Multilingual
8/10
Reasoning: Poor Coding: Weak Instruction Following: Weak Structured Output: Poor Math: Poor Multilingual: Strong

qwen3-30b-a3b-thinking-…

Reasoning
19/20
Coding
18/20
Instruction
19/20
Structured
15/15
Math
15/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Strong Multilingual: Strong
View deepseek-r1-distil… details | View qwen3-30b-a3b-thin… 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