richardyoung/qwythos-9b-abliterated:Q8_0
OLLAMA GGUFqwen35 · 9.0B · Q8_0
Jun 28, 2026 · Apple M4 Pro
llama3.1:8b
OLLAMA GGUFllama · 8.0B · Q4_K_M
Sep 15, 2026 · Intel Core™ i5-14400F
Global Score
65 vs 81
Hardware Fit
96 vs 100
Quality Score
52 vs 73
Hardware
richardyoung/qwyth… llama3.1:8b
MachineMac miniGigabyte Technology Co., Ltd. B760 AORUS ELITE AX DDR4
CPUApple M4 ProIntel Core™ i5-14400F
Cores1216
RAM24 GB32 GB
GPUApple M4 ProNVIDIA GeForce RTX 5060 Ti
OSmacOS 15.6.1Microsoft Windows 11 Professionnel 10.0.26200
Archarm64x64
Power Modebalancedbalanced
Performance
richardyoung/qwyth… llama3.1:8b
Tokens/sec25.370.2
First chunk397 ms58 ms
TTFT397 ms58 ms
Load time6.3 s4.1 s
Memory usage8.3 GB4.9 GB
Memory %34%15%
HW Fit Score Breakdown
richardyoung/qwythos-9b…
Speed
47/50
TTFT
20/20
Memory
29/30
llama3.1:8b
Speed
50/50
TTFT
20/20
Memory
30/30
Quality
richardyoung/qwythos-9b…
Reasoning
18/20
Coding
13/20
Instruction
6/20
Structured
0/15
Math
14/15
Multilingual
1/10
Reasoning: Strong Coding: Adequate Instruction Following: Weak Structured Output: Poor Math: Strong Multilingual: Poor
llama3.1:8b
Reasoning
12/20
Coding
18/20
Instruction
15/20
Structured
13/15
Math
5/15
Multilingual
10/10
Reasoning: Adequate 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