deepseek-r1:1.5b

qwen2 · 1.8B · Q4_K_M

THINKING MODEL ECO MODE

LENOVO 20L7CTO1WW (Intel Core™ i5-8350U)

23 GB · Fedora Linux 43

Tested on April 29, 2026

Top 93% Compare

Global Score

37 /100

Not Rec.

Hardware Fit

73/100

Quality

22/100

Get this model

🦙

Ollama

ollama pull deepseek-r1:1.5b

View on Ollama Library

ollama.com/library/deepseek-r1

Get it in LM Studio

Search and download models directly from the app

🤗

Find on HuggingFace

GGUF versions & quantizations

Hardware

Machine: LENOVO 20L7CTO1WW
CPU: Intel Core™ i5-8350U
Cores: 8 threads (4 cores)
Frequency: 1.7 GHz
RAM: 23 GB
GPU: Kaby Lake-R GT2 [UHD Graphics 620]
OS: Fedora Linux 43
Arch: x64
Power Mode: low-power

Performance

Tokens/sec Tokens generated per second — higher is better: 18.4
Standard deviation Variation of speed across benchmark runs — lower means more consistent: ±2.1
First chunk latency Delay before the first streamed chunk arrives from runtime: 593 ms
Time to first token How fast the model starts responding — lower is better: 30.0 s
Load time Time to load the model into memory before inference: 3.8 s
Memory usage RAM consumed during inference — percentage of total system memory: 1.0 GB (4%)
Total tokens Total number of tokens generated across all benchmark prompts: 1394
Thinking tokens (est.) Estimated internal reasoning tokens — models with "thinking" generate more tokens: ~1055

Score breakdown

Speed

43/50

Time to first token

0/20

Memory

30/30

Quality

Reasoning

4/20

Coding

0/20

Instruction following

5/20

Structured output

3/15

Math

7/15

Multilingual

3/10

Category levels

Reasoning: Poor Coding: Poor Instruction Following: Weak Structured Output: Poor Math: Weak Multilingual: Weak

Metadata

Spec version: 0.2.1
Runtime: Ollama 0.21.2
Model format: GGUF
Hardware profile: ENTRY
Result hash: 6627bd289e44e5620482bad9eb1701fcc1a418a70b700b38d469bc0390f8dd27

Interpretation

Hardware fit: 73/100. Overall suitability: NOT RECOMMENDED (Global 37/100). Category profile: Reasoning: Poor, Coding: Poor, Instruction Following: Weak, Structured Output: Poor, Math: Weak, Multilingual: Weak.

Warnings

System was in low-power mode during this benchmark.

Disqualifiers

Time to first token too high: 30000ms (maximum: 22196ms for ENTRY profile)

Bench Environment

Thermal: nominal Power: AC CPU load: avg 67% (peak 85%)

Run yours now

$ npm install -g metrillm@latest

$ metrillm

Requires Node 20+ and Ollama or LM Studio running

Or run without installing: npx metrillm@latest