Back to leaderboard

gemma4:latest

OLLAMA GGUF

gemma4 · 8.0B · Q4_K_M

Excellent

Jun 5, 2026 · AMD Ryzen 7 5800H

qwen3.5-35b-a3b-t48-c8k:latest

OLLAMA GGUF

qwen35moe · 36.0B · Q4_K_M

Not Rec.

Aug 1, 2026 · Intel Xeon® E5-4620 0

Global Score
89 vs 65
Hardware Fit
83 vs 38
Quality Score
92 vs 76

Hardware

gemma4:latest qwen3.5-35b-a3b-t4…
MachineLENOVO 82JQDell Inc. PowerEdge R820
CPUAMD Ryzen 7 5800HIntel Xeon® E5-4620 0
Cores1664
RAM16 GB256 GB
GPUNVIDIA GeForce RTX 3060 Laptop GPUMicrosoft Remote Display Adapter, Microsoft Basic Display Adapter
OSMicrosoft Windows 11 家庭版 中文版 10.0.26100Microsoft Windows Server 2022 Datacenter 10.0.20348
Archx64x64
Power Modebalancedbalanced

Performance

gemma4:latest qwen3.5-35b-a3b-t4…
Tokens/sec46.33.4
First chunk824 ms4210 ms
TTFT3.8 s4.2 s
Load time21.0 s45.1 s
Memory usage8.9 GB22.0 GB
Memory %56%9%

HW Fit Score Breakdown

gemma4:latest

Speed
50/50
TTFT
11/20
Memory
22/30

qwen3.5-35b-a3b-t48-c8k…

Speed
3/50
TTFT
5/20
Memory
30/30

Quality

gemma4:latest

Reasoning
18/20
Coding
19/20
Instruction
15/20
Structured
15/15
Math
15/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Strong Multilingual: Strong

qwen3.5-35b-a3b-t48-c8k…

Reasoning
16/20
Coding
15/20
Instruction
9/20
Structured
15/15
Math
11/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Weak Structured Output: Strong Math: Adequate 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