qwen3:30b-a3b-instruct-2507-q4_K_M
OLLAMA GGUFqwen3moe · 30.5B · Q4_K_M
Jul 31, 2026 · Intel Xeon® E5-4620 0
deepseek-r1:1.5b
OLLAMA GGUFqwen2 · 1.8B · Q4_K_M
Apr 29, 2026 · Intel Core™ i5-8350U
Global Score
71 vs 37
Hardware Fit
42 vs 73
Quality Score
84 vs 22
Hardware
qwen3:30b-a3b-inst… deepseek-r1:1.5b
MachineDell Inc. PowerEdge R820LENOVO 20L7CTO1WW
CPUIntel Xeon® E5-4620 0Intel Core™ i5-8350U
Cores648
RAM256 GB23 GB
GPUMicrosoft Remote Display Adapter, Microsoft Basic Display AdapterKaby Lake-R GT2 [UHD Graphics 620]
OSMicrosoft Windows Server 2022 Datacenter 10.0.20348Fedora Linux 43
Archx64x64
Power Modebalancedlow-power
Performance
qwen3:30b-a3b-inst… deepseek-r1:1.5b
Tokens/sec4.018.4
First chunk2070 ms593 ms
TTFT2.1 s30.0 s
Load time44.5 s3.8 s
Memory usage17.7 GB1.0 GB
Memory %7%4%
HW Fit Score Breakdown
qwen3:30b-a3b-instruct-…
Speed
3/50
TTFT
9/20
Memory
30/30
deepseek-r1:1.5b
Speed
43/50
TTFT
0/20
Memory
30/30
Quality
qwen3:30b-a3b-instruct-…
Reasoning
18/20
Coding
18/20
Instruction
11/20
Structured
15/15
Math
12/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Adequate Structured Output: Strong Math: Strong Multilingual: Strong
deepseek-r1:1.5b
Reasoning
4/20
Coding
0/20
Instruction
5/20
Structured
3/15
Math
7/15
Multilingual
3/10
Reasoning: Poor Coding: Poor Instruction Following: Weak Structured Output: Poor Math: Weak Multilingual: Weak
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