prism-ml/bonsai-27b

qwen35 · 27B · Q1_0

ASUSTeK COMPUTER INC. ROG Strix G16 G614PH_G614PH (AMD Ryzen 9 8940HX)

15 GB · Microsoft Windows 11 Home Single Language 10.0.26200

Tested on September 20, 2026

Top 38% Compare

Global Score

81 /100

Excellent

Hardware Fit

95/100

Quality

75/100

Get this model

Get it in LM Studio

Search and download models directly from the app

🤗

Find on HuggingFace

MLX versions & conversions

Hardware

Machine: ASUSTeK COMPUTER INC. ROG Strix G16 G614PH_G614PH
CPU: AMD Ryzen 9 8940HX
Cores: 32 total (32 perf)
Frequency: 2.4 GHz
RAM: 15 GB DDR5
GPU: NVIDIA GeForce RTX 5050 Laptop GPU, AMD Radeon(TM) 610M
OS: Microsoft Windows 11 Home Single Language 10.0.26200
Arch: x64
Power Mode: balanced

Performance

Tokens/sec Tokens generated per second — higher is better: 33.6
Standard deviation Variation of speed across benchmark runs — lower means more consistent: ±0.7
First chunk latency Delay before the first streamed chunk arrives from runtime: 20 ms
Time to first token How fast the model starts responding — lower is better: 409 ms
Load time Time to load the model into memory before inference: N/A
Memory usage RAM consumed during inference — percentage of total system memory: 5.3 GB (35%)
Total tokens Total number of tokens generated across all benchmark prompts: 1290

Score breakdown

Speed

46/50

Time to first token

20/20

Memory

29/30

Quality

Reasoning

17/20

Coding

13/20

Instruction following

13/20

Structured output

15/15

Math

8/15

Multilingual

9/10

Category levels

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

Metadata

Spec version: 0.2.1
Runtime: LM Studio
Model format: GGUF
Hardware profile: BALANCED
Result hash: 30837c4120077ccba6a89d5b667c505a0c6090b24c57bc120d80b40f377a623d

Interpretation

Hardware fit: 95/100. Overall suitability: EXCELLENT (Global 81/100). Category profile: Reasoning: Strong, Coding: Adequate, Instruction Following: Adequate, Structured Output: Strong, Math: Adequate, Multilingual: Strong.

Warnings

Model memory footprint is estimated via LM Studio CLI rather than measured from a fresh load.
Thermal throttling detected after benchmark (heavy). Results may be degraded — consider cooling or shorter runs.

Bench Environment

Thermal: heavy Power: AC CPU load: avg 7% (peak 10%)

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