Back to leaderboard

qwen3-coder-30b-a3b-instruct-mlx

LM-STUDIO MLX

qwen3_moe · 30B · 4bit

Excellent

Oct 20, 2026 · Apple M1 Pro

unsloth/gemma-4-12b-it-qat

LM-STUDIO GGUF

gemma4 · 12B · Q4_K_XL

Good

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

Global Score
87 vs 75
Hardware Fit
85 vs 48
Quality Score
88 vs 86

Hardware

qwen3-coder-30b-a3… unsloth/gemma-4-12…
MachineMacBook ProDocker Container
CPUApple M1 ProIntel Gen Intel® Core™ i9-11900H
Cores1016
RAM32 GB15 GB
GPUApple M1 ProTigerLake-H GT1 [UHD Graphics]
OSmacOS 15.6.1Ubuntu 26.04 LTS
Archarm64x64
Power Modebalancedperformance

Performance

qwen3-coder-30b-a3… unsloth/gemma-4-12…
Tokens/sec45.76.0
First chunk17 ms27 ms
TTFT390 ms5.1 s
Load timeN/A8.4 s
Memory usage22.4 GB0.0 GB
Memory %70%0%

HW Fit Score Breakdown

qwen3-coder-30b-a3b-ins…

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

unsloth/gemma-4-12b-it-…

Speed
10/50
TTFT
8/20
Memory
30/30

Quality

qwen3-coder-30b-a3b-ins…

Reasoning
15/20
Coding
19/20
Instruction
17/20
Structured
15/15
Math
12/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Strong Multilingual: Strong

unsloth/gemma-4-12b-it-…

Reasoning
16/20
Coding
18/20
Instruction
15/20
Structured
15/15
Math
12/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Strong Multilingual: Strong
View qwen3-coder-30b-a3… details | View unsloth/gemma-4-12… 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