Back to leaderboard

qwen3-30b-a3b-thinking-2507-claude-4.5-sonnet-high-reasoning-distill-mlx

LM-STUDIO MLX

qwen3_moe · 30B · MXFP4

Excellent

Apr 22, 2026 · Apple M2 Max

hf.co/empero-ai/Qwen3.8-4B-Distill-GGUF:Q4_K_M

OLLAMA GGUF

qwen35 · 4.33B · unknown

Good

Aug 28, 2026 · Apple M4

Global Score
95 vs 66
Hardware Fit
94 vs 100
Quality Score
96 vs 52

Hardware

qwen3-30b-a3b-thin… hf.co/empero-ai/Qw…
MachineMac StudioMacBook Pro
CPUApple M2 MaxApple M4
Cores1210
RAM96 GB16 GB
GPUApple M2 MaxApple M4
OSmacOS 26.1macOS 26.6.1
Archarm64arm64
Power Modebalancedbalanced

Performance

qwen3-30b-a3b-thin… hf.co/empero-ai/Qw…
Tokens/sec80.426.5
First chunk46 ms447 ms
TTFT1.8 s447 ms
Load timeN/A3.3 s
Memory usage21.2 GB4.8 GB
Memory %22%30%

HW Fit Score Breakdown

qwen3-30b-a3b-thinking-…

Speed
50/50
TTFT
14/20
Memory
30/30

hf.co/empero-ai/Qwen3.8…

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

hf.co/empero-ai/Qwen3.8…

Reasoning
18/20
Coding
14/20
Instruction
4/20
Structured
1/15
Math
15/15
Multilingual
0/10
Reasoning: Strong Coding: Adequate Instruction Following: Poor Structured Output: Poor Math: Strong Multilingual: Poor
View qwen3-30b-a3b-thin… details | View hf.co/empero-ai/Qw… 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