Back to leaderboard

gpt-oss:20b

OLLAMA GGUF

gptoss · 20.9B · MXFP4

Excellent

Jul 3, 2026 · AMD RYZEN AI MAX+ 395

qwen2.5:72b

OLLAMA GGUF

qwen2 · 72.7B · Q4_K_M

Not Rec.

Jun 27, 2026 · Cortex-X925

Global Score
94 vs 75
Hardware Fit
91 vs 50
Quality Score
95 vs 86

Hardware

gpt-oss:20b qwen2.5:72b
MachineFramework Desktop (AMD Ryzen AI Max 300 Series)NVIDIA NVIDIA_DGX_Spark
CPUAMD RYZEN AI MAX+ 395Cortex-X925
Cores3220
RAM64 GB122 GB
GPUAMD Radeon(TM) 8060S GraphicsDevice 2e12
OSMicrosoft Windows 11 Pro for Workstations 10.0.26200Ubuntu 24.04.4 LTS
Archx64arm64
Power Modebalancedperformance

Performance

gpt-oss:20b qwen2.5:72b
Tokens/sec48.04.4
First chunk357 ms529 ms
TTFT2.3 s529 ms
Load time0.2 s49.7 s
Memory usage16.4 GB54.0 GB
Memory %26%44%

HW Fit Score Breakdown

gpt-oss:20b

Speed
50/50
TTFT
11/20
Memory
30/30

qwen2.5:72b

Speed
4/50
TTFT
20/20
Memory
26/30

Quality

gpt-oss:20b

Reasoning
19/20
Coding
19/20
Instruction
17/20
Structured
15/15
Math
15/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Strong Multilingual: Strong

qwen2.5:72b

Reasoning
19/20
Coding
19/20
Instruction
12/20
Structured
14/15
Math
12/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Adequate Structured Output: Strong Math: Strong Multilingual: Strong

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