Back to leaderboard

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

LM-STUDIO MLX

qwen2 · 14B · 5bit

Not Rec.

Mar 11, 2026 · Apple M4

unsloth/gemma-4-26b-a4b-it-qat

LM-STUDIO GGUF

gemma4 · 26B · Q4_K_XL

Excellent

Jun 29, 2026 · Intel Gen Intel® Core™ i9-11900H

Global Score
36 vs 85
Hardware Fit
53 vs 65
Quality Score
28 vs 93

Hardware

deepseek-r1-distil… unsloth/gemma-4-26…
MachineMac miniDocker Container
CPUApple M4Intel Gen Intel® Core™ i9-11900H
Cores1016
RAM24 GB15 GB
GPUApple M4TigerLake-H GT1 [UHD Graphics]
OSmacOS 26.3Ubuntu 26.04 LTS
Archarm64x64
Power Modebalancedperformance

Performance

deepseek-r1-distil… unsloth/gemma-4-26…
Tokens/sec10.311.9
First chunk27 ms17 ms
TTFT20.7 s3.3 s
Load time15.4 s69.8 s
Memory usage1.0 GB0.5 GB
Memory %4%3%

HW Fit Score Breakdown

deepseek-r1-distill-qwe…

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

unsloth/gemma-4-26b-a4b…

Speed
22/50
TTFT
13/20
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

unsloth/gemma-4-26b-a4b…

Reasoning
19/20
Coding
19/20
Instruction
19/20
Structured
14/15
Math
13/15
Multilingual
9/10
Reasoning: Strong Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Strong Multilingual: Strong
View deepseek-r1-distil… details | View unsloth/gemma-4-26… 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