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
qwen3:8b
OLLAMA GGUFqwen3 · 8.2B · Q4_K_M
Mar 5, 2026 · Intel Core™ Ultra 9 285K
Global Score
95 vs 84
Hardware Fit
94 vs 97
Quality Score
96 vs 79
Hardware
qwen3-30b-a3b-thin… qwen3:8b
MachineMac StudioASUS
CPUApple M2 MaxIntel Core™ Ultra 9 285K
Cores1224
RAM96 GB47 GB
GPUApple M2 MaxNVIDIA GeForce RTX 5090, Intel(R) Graphics
OSmacOS 26.1Microsoft Windows 11 Pro 10.0.26200
Archarm64x64
Power Modebalancedbalanced
Performance
qwen3-30b-a3b-thin… qwen3:8b
Tokens/sec80.4208.0
First chunk46 ms82 ms
TTFT1.8 s1.3 s
Load timeN/A1.5 s
Memory usage21.2 GB9.6 GB
Memory %22%20%
HW Fit Score Breakdown
qwen3-30b-a3b-thinking-…
Speed
50/50
TTFT
14/20
Memory
30/30
qwen3:8b
Speed
50/50
TTFT
18/20
Memory
29/30
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
qwen3:8b
Reasoning
18/20
Coding
10/20
Instruction
14/20
Structured
15/15
Math
12/15
Multilingual
10/10
Reasoning: Strong Coding: Weak Instruction Following: Adequate 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