Back to leaderboard

Qwen3.8-27B-IQ3_S-3.23bpw:latest

OLLAMA GGUF

qwen35 · 27.3B · unknown

Good

Sep 16, 2026 · AMD Ryzen 7 9700X 8-Core Processor

gpt-oss-safeguard-20b-mlx

LM-STUDIO MLX

gpt_oss · 20B · MXFP4

Excellent

Mar 12, 2026 · Apple M4

Global Score
75 vs 87
Hardware Fit
99 vs 82
Quality Score
64 vs 89

Hardware

Qwen3.8-27B-IQ3_S-… gpt-oss-safeguard-…
MachineMicro-Star International Co., Ltd. MS-7E51Mac mini
CPUAMD Ryzen 7 9700X 8-Core ProcessorApple M4
Cores1610
RAM31 GB24 GB
GPUNVIDIA GeForce RTX 5080, AMD Radeon(TM) GraphicsApple M4
OSMicrosoft Windows 11 Pro 10.0.26200macOS 26.3
Archx64arm64
Power Modebalancedbalanced

Performance

Qwen3.8-27B-IQ3_S-… gpt-oss-safeguard-…
Tokens/sec99.941.4
First chunk212 ms8 ms
TTFT212 ms3.6 s
Load time0.0 sN/A
Memory usage10.5 GB14.5 GB
Memory %34%61%

HW Fit Score Breakdown

Qwen3.8-27B-IQ3_S-3.23b…

Speed
50/50
TTFT
20/20
Memory
29/30

gpt-oss-safeguard-20b-mlx

Speed
50/50
TTFT
12/20
Memory
20/30

Quality

Qwen3.8-27B-IQ3_S-3.23b…

Reasoning
19/20
Coding
18/20
Instruction
7/20
Structured
0/15
Math
14/15
Multilingual
6/10
Reasoning: Strong Coding: Strong Instruction Following: Weak Structured Output: Poor Math: Strong Multilingual: Adequate

gpt-oss-safeguard-20b-mlx

Reasoning
19/20
Coding
12/20
Instruction
19/20
Structured
15/15
Math
15/15
Multilingual
9/10
Reasoning: Strong Coding: Adequate Instruction Following: Strong Structured Output: Strong Math: Strong Multilingual: Strong
View Qwen3.8-27B-IQ3_S-… details | View gpt-oss-safeguard-… details

Run yours and compare

$ npm install -g metrillm@latest
$ metrillm

Requires Node 20+ and Ollama or LM Studio running

Or run without installing: npx metrillm@latest