Back to leaderboard

rafw007/qwen3.8-27b-redteam:latest

OLLAMA GGUF

qwen35 · 27.3B · Q4_K_M

Marginal

Oct 2, 2026 · Cortex-X925

baidu/ernie-4.5-21b-a3b

LM-STUDIO GGUF

ernie4_5-moe · 21B · Q4_K_M

Excellent

Nov 1, 2026 · AMD RYZEN AI MAX+ 395

Global Score
59 vs 86
Hardware Fit
62 vs 93
Quality Score
58 vs 83

Hardware

rafw007/qwen3.8-27… baidu/ernie-4.5-21…
MachineNVIDIA NVIDIA_DGX_SparkGMKtec EVO-X3
CPUCortex-X925AMD RYZEN AI MAX+ 395
Cores2032
RAM122 GB120 GB
GPUDevice 2e12AMD Radeon(TM) 8060S Graphics
OSUbuntu 24.04.5 LTSMicrosoft Windows 11 Pro 10.0.26300
Archarm64x64
Power Modeperformancebalanced

Performance

rafw007/qwen3.8-27… baidu/ernie-4.5-21…
Tokens/sec12.946.3
First chunk442 ms33 ms
TTFT442 ms383 ms
Load time6.8 s49.7 s
Memory usage19.3 GB32.5 GB
Memory %16%27%

HW Fit Score Breakdown

rafw007/qwen3.8-27b-red…

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

baidu/ernie-4.5-21b-a3b

Speed
43/50
TTFT
20/20
Memory
30/30

Quality

rafw007/qwen3.8-27b-red…

Reasoning
18/20
Coding
15/20
Instruction
7/20
Structured
1/15
Math
14/15
Multilingual
3/10
Reasoning: Strong Coding: Adequate Instruction Following: Weak Structured Output: Poor Math: Strong Multilingual: Weak

baidu/ernie-4.5-21b-a3b

Reasoning
16/20
Coding
17/20
Instruction
15/20
Structured
14/15
Math
12/15
Multilingual
9/10
Reasoning: Strong Coding: Strong Instruction Following: Strong 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