Back to leaderboard

qwen3-235b-a22b-t48-c8k:latest

OLLAMA GGUF

qwen3moe · 235.1B · Q4_K_M

Not Rec.

Aug 2, 2026 · Intel Xeon® E5-4620 0

gemma3:4b

OLLAMA GGUF

gemma3 · 4.3B · Q4_K_M

Good

Jun 26, 2026 · Cortex-A78AE

Global Score
43 vs 76
Hardware Fit
28 vs 92
Quality Score
49 vs 69

Hardware

qwen3-235b-a22b-t4… gemma3:4b
MachineDell Inc. PowerEdge R820NVIDIA NVIDIA Jetson Orin Nano Engineering Reference Developer Kit Super
CPUIntel Xeon® E5-4620 0Cortex-A78AE
Cores646
RAM256 GB7 GB
GPUMicrosoft Remote Display Adapter, Microsoft Basic Display AdapterIntegrated / Unknown
OSMicrosoft Windows Server 2022 Datacenter 10.0.20348Ubuntu 22.04.5 LTS
Archx64arm64
Power Modebalancedbalanced

Performance

qwen3-235b-a22b-t4… gemma3:4b
Tokens/sec0.715.0
First chunk6361 ms2380 ms
TTFT6.4 s2.4 s
Load time0.9 s13.6 s
Memory usage134.2 GB2.6 GB
Memory %52%35%

HW Fit Score Breakdown

qwen3-235b-a22b-t48-c8k…

Speed
1/50
TTFT
3/20
Memory
24/30

gemma3:4b

Speed
45/50
TTFT
18/20
Memory
29/30

Quality

qwen3-235b-a22b-t48-c8k…

Reasoning
19/20
Coding
2/20
Instruction
4/20
Structured
2/15
Math
13/15
Multilingual
9/10
Reasoning: Strong Coding: Poor Instruction Following: Poor Structured Output: Poor Math: Strong Multilingual: Strong

gemma3:4b

Reasoning
11/20
Coding
17/20
Instruction
13/20
Structured
14/15
Math
4/15
Multilingual
10/10
Reasoning: Adequate Coding: Strong Instruction Following: Adequate Structured Output: Strong Math: Weak 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