qwen3:4b
OLLAMA GGUFqwen3 · 4.0B · Q4_K_M
Nov 4, 2026 · Intel Core™ i7-9700
qwen3.8-27b-mlx
LM-STUDIO MLXqwen3_5 · 27B · 4bit
Oct 20, 2026 · Apple M1 Pro
Global Score
29 vs 78
Hardware Fit
84 vs 56
Quality Score
6 vs 87
Hardware
qwen3:4b qwen3.8-27b-mlx
MachineVMware, Inc. VMware Virtual PlatformMacBook Pro
CPUIntel Core™ i7-9700Apple M1 Pro
Cores410
RAM16 GB32 GB
GPUIntegrated / UnknownApple M1 Pro
OSUbuntu 24.04.5 LTSmacOS 15.6.1
Archx64arm64
Power Modeunknownbalanced
Performance
qwen3:4b qwen3.8-27b-mlx
Tokens/sec10.410.8
First chunk857 ms8 ms
TTFT857 ms2.6 s
Load time41.0 sN/A
Memory usage3.0 GB21.0 GB
Memory %19%66%
HW Fit Score Breakdown
qwen3:4b
Speed
34/50
TTFT
20/20
Memory
30/30
qwen3.8-27b-mlx
Speed
23/50
TTFT
16/20
Memory
17/30
Quality
qwen3:4b
Reasoning
0/20
Coding
2/20
Instruction
2/20
Structured
0/15
Math
1/15
Multilingual
1/10
Reasoning: Poor Coding: Poor Instruction Following: Poor Structured Output: Poor Math: Poor Multilingual: Poor
qwen3.8-27b-mlx
Reasoning
18/20
Coding
18/20
Instruction
15/20
Structured
15/15
Math
11/15
Multilingual
10/10
Reasoning: Strong 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