hf.co/mradermacher/gemma-4-E4B-GGUF:Q6_K
OLLAMA GGUFgemma4 · 7.52B · unknown
Aug 24, 2026 · Apple M1 Pro
qwen3-30b-a3b-thinking-2507-claude-4.5-sonnet-high-reasoning-distill-mlx
LM-STUDIO MLXqwen3_moe · 30B · MXFP4
Apr 22, 2026 · Apple M2 Max
Global Score
41 vs 95
Hardware Fit
100 vs 94
Quality Score
16 vs 96
Hardware
hf.co/mradermacher… qwen3-30b-a3b-thin…
MachineMacBook ProMac Studio
CPUApple M1 ProApple M2 Max
Cores1012
RAM16 GB96 GB
GPUApple M1 ProApple M2 Max
OSmacOS 26.6.2macOS 26.1
Archarm64arm64
Power Modebalancedbalanced
Performance
hf.co/mradermacher… qwen3-30b-a3b-thin…
Tokens/sec34.680.4
First chunk179 ms46 ms
TTFT179 ms1.8 s
Load time3.9 sN/A
Memory usage3.9 GB21.2 GB
Memory %25%22%
HW Fit Score Breakdown
hf.co/mradermacher/gemm…
Speed
50/50
TTFT
20/20
Memory
30/30
qwen3-30b-a3b-thinking-…
Speed
50/50
TTFT
14/20
Memory
30/30
Quality
hf.co/mradermacher/gemm…
Reasoning
5/20
Coding
7/20
Instruction
0/20
Structured
0/15
Math
3/15
Multilingual
1/10
Reasoning: Poor Coding: Weak Instruction Following: Poor Structured Output: Poor Math: Poor Multilingual: Poor
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
Run yours and compare
$
npm install -g metrillm@latest$
metrillmRequires Node 20+ and Ollama or LM Studio running
Or run without installing: npx metrillm@latest