mistralai/ministral-3-3b
LM-STUDIO GGUFmistral3 · 3B · Q4_K_M
Oct 31, 2026 · AMD RYZEN AI MAX+ 395
qwen3.5-2b-mlx
LM-STUDIO MLXqwen3_5 · 2B · 8bit
Mar 3, 2026 · Apple M4
Global Score
76 vs 70
Hardware Fit
94 vs 93
Quality Score
68 vs 60
Hardware
mistralai/ministra… qwen3.5-2b-mlx
MachineGMKtec EVO-X3MacBook Air
CPUAMD RYZEN AI MAX+ 395Apple M4
Cores3210
RAM120 GB32 GB
GPUAMD Radeon(TM) 8060S GraphicsApple M4
OSMicrosoft Windows 11 Pro 10.0.26300macOS 26.3
Archx64arm64
Power Modebalancedbalanced
Performance
mistralai/ministra… qwen3.5-2b-mlx
Tokens/sec50.232.9
First chunk6 ms3096 ms
TTFT726 ms3.1 s
Load timeN/AN/A
Memory usage19.6 GB2.5 GB
Memory %16%8%
HW Fit Score Breakdown
mistralai/ministral-3-3b
Speed
46/50
TTFT
18/20
Memory
30/30
qwen3.5-2b-mlx
Speed
50/50
TTFT
13/20
Memory
30/30
Quality
mistralai/ministral-3-3b
Reasoning
10/20
Coding
16/20
Instruction
15/20
Structured
15/15
Math
3/15
Multilingual
9/10
Reasoning: Adequate Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Poor Multilingual: Strong
qwen3.5-2b-mlx
Reasoning
11/20
Coding
9/20
Instruction
14/20
Structured
14/15
Math
3/15
Multilingual
9/10
Reasoning: Adequate Coding: Weak Instruction Following: Adequate Structured Output: Strong Math: Poor 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