qwen3.8-27b-claude-opus-reasoning-distilled
LM-STUDIO GGUFqwen35 · 27B · Q4_K_M
Aug 20, 2026 · Apple M4 Max
ministral-3:3b
OLLAMA GGUFmistral3 · 3.8B · Q4_K_M
Mar 6, 2026 · Apple M4
Global Score
80 vs 71
Hardware Fit
63 vs 99
Quality Score
87 vs 59
Hardware
qwen3.8-27b-claude… ministral-3:3b
MachineMacBook ProMacBook Air
CPUApple M4 MaxApple M4
Cores1610
RAM128 GB32 GB
GPUApple M4 MaxApple M4
OSmacOS 26.5.2macOS 26.3
Archarm64arm64
Power Modebalancedbalanced
Performance
qwen3.8-27b-claude… ministral-3:3b
Tokens/sec24.042.2
First chunk20 ms189 ms
TTFT4.5 s189 ms
Load timeN/A1.2 s
Memory usage15.7 GB7.0 GB
Memory %12%22%
HW Fit Score Breakdown
qwen3.8-27b-claude-opus…
Speed
27/50
TTFT
6/20
Memory
30/30
ministral-3:3b
Speed
50/50
TTFT
20/20
Memory
29/30
Quality
qwen3.8-27b-claude-opus…
Reasoning
19/20
Coding
18/20
Instruction
11/20
Structured
15/15
Math
14/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Adequate Structured Output: Strong Math: Strong Multilingual: Strong
ministral-3:3b
Reasoning
5/20
Coding
16/20
Instruction
11/20
Structured
14/15
Math
4/15
Multilingual
9/10
Reasoning: Weak Coding: Strong Instruction Following: Adequate 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