qwen/qwen3.8-27b
LM-STUDIO MLXqwen3_5 · 27B · 4bit
Aug 18, 2026 · Apple M4 Max
huihui-devstral-small-2-24b-instruct-2512-abliterated-i1
LM-STUDIO GGUFmistral3 · 24B · Q4_K_S
Sep 19, 2026 · AMD Ryzen 7 5800X 8-Core Processor
Global Score
87 vs 78
Hardware Fit
83 vs 74
Quality Score
88 vs 79
Hardware
qwen/qwen3.8-27b huihui-devstral-sm…
MachineMacBook ProSystem manufacturer
CPUApple M4 MaxAMD Ryzen 7 5800X 8-Core Processor
Cores1616
RAM128 GB32 GB
GPUApple M4 MaxAMD Radeon RX 9060 XT
OSmacOS 26.5.2Microsoft Windows 11 Home 10.0.26200
Archarm64x64
Power Modebalancedbalanced
Performance
qwen/qwen3.8-27b huihui-devstral-sm…
Tokens/sec29.117.5
First chunk9 ms20 ms
TTFT650 ms279 ms
Load timeN/AN/A
Memory usage21.0 GB15.6 GB
Memory %16%49%
HW Fit Score Breakdown
qwen/qwen3.8-27b
Speed
33/50
TTFT
20/20
Memory
30/30
huihui-devstral-small-2…
Speed
29/50
TTFT
20/20
Memory
25/30
Quality
qwen/qwen3.8-27b
Reasoning
18/20
Coding
18/20
Instruction
16/20
Structured
15/15
Math
11/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Adequate Multilingual: Strong
huihui-devstral-small-2…
Reasoning
16/20
Coding
17/20
Instruction
15/20
Structured
14/15
Math
7/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Weak 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