qwen3.8:27b-bf16
OLLAMA GGUFqwen35 · 27.3B · BF16
Aug 22, 2026 · AMD EPYC 7713 64-Core Processor
gemma4:26b
OLLAMA GGUFgemma4 · 25.8B · Q4_K_M
Apr 3, 2026 · AMD RYZEN AI MAX+ 395
Global Score
85 vs 89
Hardware Fit
77 vs 95
Quality Score
88 vs 87
Hardware
qwen3.8:27b-bf16 gemma4:26b
MachineDell Inc. PowerEdge XE8545AZW GTR Pro
CPUAMD EPYC 7713 64-Core ProcessorAMD RYZEN AI MAX+ 395
Cores12832
RAM1007 GB125 GB
GPUGA100 [A100 SXM4 80GB], GA100 [A100 SXM4 80GB], Integrated Matrox G200eW3 Graphics Controller, GA100 [A100 SXM4 80GB], GA100 [A100 SXM4 80GB]Radeon 8060S
OSRocky Linux 8.10Ubuntu 24.04.4 LTS
Archx64x64
Power Modeunknownperformance
Performance
qwen3.8:27b-bf16 gemma4:26b
Tokens/sec27.248.3
First chunk237 ms346 ms
TTFT237 ms346 ms
Load time43.8 s4.4 s
Memory usage68.8 GB36.1 GB
Memory %7%29%
HW Fit Score Breakdown
qwen3.8:27b-bf16
Speed
27/50
TTFT
20/20
Memory
30/30
gemma4:26b
Speed
45/50
TTFT
20/20
Memory
30/30
Quality
qwen3.8:27b-bf16
Reasoning
18/20
Coding
18/20
Instruction
16/20
Structured
15/15
Math
11/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Adequate Multilingual: Strong
gemma4:26b
Reasoning
16/20
Coding
19/20
Instruction
16/20
Structured
14/15
Math
13/15
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$
metrillmRequires Node 20+ and Ollama or LM Studio running
Or run without installing: npx metrillm@latest