qwen3-30b-a3b-thinking-2507-claude-4.5-sonnet-high-reasoning-distill-mlx
LM-STUDIO MLXqwen3_moe · 30B · MXFP4
Aug 20, 2026 · Apple M4 Max
gemma4:e2b
OLLAMA GGUFgemma4 · 4.6B · Q4_K_M
Oct 12, 2026 · AMD Ryzen 5 5500
Global Score
96 vs 85
Hardware Fit
95 vs 100
Quality Score
96 vs 78
Hardware
qwen3-30b-a3b-thin… gemma4:e2b
MachineMacBook ProMicrosoft WSL
CPUApple M4 MaxAMD Ryzen 5 5500
Cores1612
RAM128 GB8 GB
GPUApple M4 MaxIntegrated / Unknown
OSmacOS 26.5.2Ubuntu 26.04.1 LTS
Archarm64x64
Power Modebalancedunknown
Performance
qwen3-30b-a3b-thin… gemma4:e2b
Tokens/sec112.8139.2
First chunk21 ms51 ms
TTFT1.5 s51 ms
Load time3.5 s28.5 s
Memory usage0.0 GB0.2 GB
Memory %0%3%
HW Fit Score Breakdown
qwen3-30b-a3b-thinking-…
Speed
50/50
TTFT
15/20
Memory
30/30
gemma4:e2b
Speed
50/50
TTFT
20/20
Memory
30/30
Quality
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
gemma4:e2b
Reasoning
13/20
Coding
17/20
Instruction
15/20
Structured
15/15
Math
8/15
Multilingual
10/10
Reasoning: Adequate Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Adequate 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