swift-1.5-iq2_xs

THINKING MODEL ECO MODE

Gigabyte Technology Co., Ltd. B550M AORUS PRO (AMD Ryzen 5 5600X 6-Core Processor)

47 GB · Arch Linux

Tested on November 14, 2026 · Submitted by paul

Top 0% Compare

Global Score

96 /100

Excellent

Hardware Fit

94/100

Quality

97/100

Get this model

Get it in LM Studio

Search and download models directly from the app

🤗

Find on HuggingFace

MLX versions & conversions

Hardware

Machine: Gigabyte Technology Co., Ltd. B550M AORUS PRO
CPU: AMD Ryzen 5 5600X 6-Core Processor
Cores: 12 threads (6 cores)
Frequency: 3.7 GHz
RAM: 47 GB
GPU: AD106 [GeForce RTX 4060 Ti]
OS: Arch Linux
Arch: x64
Power Mode: low-power

Performance

Tokens/sec Tokens generated per second — higher is better: 60.0
Standard deviation Variation of speed across benchmark runs — lower means more consistent: ±3.0
First chunk latency Delay before the first streamed chunk arrives from runtime: 2 ms
Time to first token How fast the model starts responding — lower is better: 2.6 s
Load time Time to load the model into memory before inference: N/A
Memory usage RAM consumed during inference — percentage of total system memory: 0.9 GB (2%)
Total tokens Total number of tokens generated across all benchmark prompts: 1288
Thinking tokens (est.) Estimated internal reasoning tokens — models with "thinking" generate more tokens: ~331

Score breakdown

Speed

50/50

Time to first token

14/20

Memory

30/30

Quality

Reasoning

20/20

Coding

19/20

Instruction following

19/20

Structured output

15/15

Math

15/15

Multilingual

9/10

Category levels

Reasoning: Strong Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Strong Multilingual: Strong

Metadata

Spec version: 0.2.1
Runtime: LM Studio
Model format: UNKNOWN
Hardware profile: BALANCED
Result hash: b2da39e7421da64363960234c0cfb9f62ba46a86f1955b776676b2697f6c2968

Interpretation

Hardware fit: 94/100. Overall suitability: EXCELLENT (Global 96/100). Category profile: Reasoning: Strong, Coding: Strong, Instruction Following: Strong, Structured Output: Strong, Math: Strong, Multilingual: Strong.

Warnings

System was in low-power mode during this benchmark.
Significant swap activity during benchmark (+0.7 GB). Model may exceed available RAM — results are severely degraded.

Bench Environment

Thermal: nominal Swap delta: +0.7 GB CPU load: avg 43% (peak 53%)

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