qwen3.8-27b-claude-opus-reasoning-distilled
LM-STUDIO GGUFqwen35 · 27B · Q4_K_M
Aug 20, 2026 · Apple M4 Max
codestral:22b
OLLAMA GGUFllama · 22.2B · Q4_0
Jun 26, 2026 · Cortex-X925
Global Score
80 vs 67
Hardware Fit
63 vs 67
Quality Score
87 vs 67
Hardware
qwen3.8-27b-claude… codestral:22b
MachineMacBook ProNVIDIA NVIDIA_DGX_Spark
CPUApple M4 MaxCortex-X925
Cores1620
RAM128 GB122 GB
GPUApple M4 MaxDevice 2e12
OSmacOS 26.5.2Ubuntu 24.04.4 LTS
Archarm64arm64
Power Modebalancedperformance
Performance
qwen3.8-27b-claude… codestral:22b
Tokens/sec24.017.2
First chunk20 ms129 ms
TTFT4.5 s129 ms
Load timeN/A11.6 s
Memory usage15.7 GB18.9 GB
Memory %12%16%
HW Fit Score Breakdown
qwen3.8-27b-claude-opus…
Speed
27/50
TTFT
6/20
Memory
30/30
codestral:22b
Speed
17/50
TTFT
20/20
Memory
30/30
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
codestral:22b
Reasoning
11/20
Coding
17/20
Instruction
9/20
Structured
14/15
Math
6/15
Multilingual
10/10
Reasoning: Adequate Coding: Strong Instruction Following: Weak 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